GDE3 FUNCTION IN OLIGODENDROCYTE PRECURSOR CELL PROLIFERATION
By Mateusz Dobrowolski
A dissertation submitted to Johns Hopkins University in conformity with the requirements for the degree of Doctor of Philosophy
Baltimore, Maryland November, 2018
Abstract
Oligodendrocyte specification, proliferation and differentiation are tightly controlled by extrinsic signals; however, proteins that modulate cellular response to these factors remain unclear. Six-transmembrane GDEs are emerging as central regulators of cellular differentiation via their unique ability to shed GPI-anchored proteins from the cell surface. We show here that GDE3 controls the pace of oligodendrocyte generation by negatively regulating oligodendrocyte precursor cell
(OPC) proliferation. GDE3 inhibits OPC proliferation by stimulating CNTF-mediated signaling through release of CNTFRα, the ligand-binding component of the gp130/LIFRβ CNTF receptor complex that can function as a soluble factor. GDE3 releases CNTFRα by GPI-anchor cleavage from the plasma membrane and from extracellular vesicles (EVs) after co-recruitment of CNTFRα in EVs. These studies uncover new physiological roles for GDE3 in gliogenesis and identify GDE3 as a key regulator of CNTF-dependent inhibition of OPC proliferation through bi-modal release of
CNTFRα.
Thesis Advisor: Dr. Shanthini Sockanathan, Ph.D, Professor of Neuroscience, The
Solomon H. Snyder Department of Neuroscience, Johns Hopkins University School of
Medicine
Reader: Dr. Alex Kolodkin, Ph.D, Professor of Neuroscience, The Solomon H. Snyder
Department of Neuroscience, Johns Hopkins University School of Medicine
ii
Table of Contents
Abstract ...... ii
Table of Contents ...... iii
List of Tables ...... vii
List of Figures ...... viii
Chapter 1 Introduction ...... 1
1.1 CNS cell diversity and oligodendrocyte development...... 2
1.1.1 Development of CNS ...... 2
1.1.2 Oligodendrocyte development ...... 3
1.1.2.1 Oligodendrocyte Proliferation ...... 4
1.1.2.2 Oligodendrocyte Migration ...... 5
1.1.2.3 Oligodendrocyte Differentiation ...... 5
1.1.3 Gp130 family of cytokines in CNS development ...... 7
1.1.3.1 Gp130 signaling in neuroepithelial precursor cells ...... 8
1.1.3.2 Gp130 signaling in glia ...... 9
1.2 GDEs – Six-transmembrane domain GDPD containing proteins ...... 10
1.2.1 GDE2...... 10
1.2.2 GDE3...... 11
1.3 Extracellular vesicles ...... 12
1.3.1 1.3.1 Biogenesis of EV ...... 13
1.3.2 Release mechanisms of EV ...... 14
1.3.3 EV content and signaling ...... 14
iii
1.4 Summary and Specific Aims ...... 16
Chapter 2 GDE3 regulates OPC proliferation and differentiation in vivo ...... 22
2.1 Introduction ...... 23
2.2 Results ...... 23
2.2.1 GDE3 regulates cell cycle kinetics of ventral progenitors in spinal cord
...... 23
2.2.2 Loss of Gde3 results in increased OPC production and differentiation in
vivo ...... 24
2.3 Summary...... 26
Chapter 3 GDE3 regulates proliferation and differentiation of purified OPCs ...... 34
3.1 Introduction ...... 35
3.2 Results ...... 35
3.2.1 GDE3 is necessary to inhibit OPC proliferation ...... 35
3.2.2 GDE3 controls the maturation of OPCs ...... 36
3.2.3 GDE3 is sufficient to inhibit OPC proliferation ...... 36
3.3 Summary...... 37
Chapter 4 Identifying pathways that mediate GDE3-dependent regulation of OPC proliferation ...... 41
4.1 Introduction ...... 42
4.2 Results ...... 42
4.2.1 RNA-seq approach ...... 42
4.2.2 Functional annotation of differentially enriched genes ...... 43
iv
4.2.3 CNTFRα, a candidate GPI-AP mediating cytokine signaling ...... 44
4.3 Summary...... 45
Chapter 5 GDE3 regulates CNTF-CNTFRα signaling in proliferating OPCs ...... 52
5.1 Introduction ...... 53
5.2 Results ...... 53
5.2.1 CNTF inhibits OPC proliferation in a GDE3 dependent manner ...... 53
5.2.2 Neutralization of CNTFRα-GP130 signaling increases OPC
proliferation ...... 54
5.3 Summary...... 55
Chapter 6 GDE3 mediates bimodal release of CNTFRα ...... 58
6.1 Introduction ...... 59
6.2 Results ...... 59
6.2.1 GDE3 releases CNTFRα via two separate mechanisms ...... 59
6.2.2 Specific domains of GDE3 regulate CNTFRα release and other GPI-APs
...... 61
6.2.3 GDE3’s release of CNTFRα correlates with inhibition of OPC
proliferation ...... 63
6.2.4 GDE3 is present on EVs and is enzymatically active ...... 64
6.2.5 Exogenous CNTFRα rescues hyperproliferation defects of Gde3 KO
OPCs ...... 65
6.3 Summary...... 66
Chapter 7 Summary and Future Directions ...... 76
v
7.1 Summary...... 77
7.2 Discussion and Future Directions ...... 79
7.2.1 GDE3 function in OPC proliferation and differentiation ...... 79
7.2.2 GDE3 regulation of CNTF signaling ...... 80
7.2.3 GDE3 and EV release ...... 83
Materials and Methods ...... 87
References ...... 95
Biography ...... 108
vi
List of Tables
Table 1.1 gp130 Cytokines and their receptor components...... …....21
Table 4.1 GPI-APs with cytokine signaling ...... …....49
Table 4.2 RNA-sequencing genes that were significantly altered in Gde3 KO OPCs
...... …....51
Table 7.1 GPS-Lipid analysis of GDE3 ...... …....86
vii
List of Figures
Figure 1.1 Markers of oligodendrocyte lineage cells through specification and maturation...... 17
Figure 1.2 Signaling pathways implicated in CNTFRα signaling ...... 18
Figure 1.3 Molecular machineries of extracellular vesicle biogenesis...... 19
Figure 1.4 GDE topology and expression of GDE3 in the developing spinal cord.
...... 20
Figure 2.1 Deletion of Gde3 increases proliferation of VZ progenitors ...... 28
Figure 2.2 Gde3 null OPCs exhibit heightened proliferation ...... 30
Figure 2.3 Gde3 regulates OPC proliferation and differentiation without changes in cell death...... 31
Figure 2.4 Gde3 regulates maturation of oligodendrocytes ...... 32
Figure 2.5 Gde3 null oligodendrocytes display increased proliferation ...... 33
Figure 3.1 Spinal cord and cortical OPCs lacking Gde3 display heightened proliferation ...... 38
Figure 3.2 Gde3 null OPCs display increased maturation in vitro ...... 39
Figure 3.3 Expression of GDE3 inhibits OPC proliferation ...... 40
Figure 4.1 RNAseq results from Gde3 KO OPCs ...... 46
Figure 4.2 OPCs express GDE3 and CNTFRα ...... 47
Figure 4.3 Levels of soluble CNTFRα are decreased in Gde3 KO animals ...... 48
Figure 5.1 OPCs require GDE3 for CNTF-mediated suppression of proliferation . 56
Figure 5.2 Neutralizing antibodies against gp130 increase WT OPC proliferation 57
Figure 6.1 GDE3 releases two species of CNTFRα...... 67
viii
Figure 6.2 Mutant versions of GDE3 are effectively processed and trafficked to the cell surface...... 68
Figure 6.3 GDE3 releases soluble and membrane bound CNTFRα...... 69
Figure 6.4 GDE3 release of GPI-APs...... 70
Figure 6.5 Evaluation of GDE3 mediated release of extracellular particles ...... 71
Figure 6.6 Proliferation assay of transfected OPCs with GDE3 constructs ...... 72
Figure 6.7 Non-membrane bound CNTFRα is sufficient to inhibit OPC proliferation...... 74
Figure 6.8 Working model for GDE3 dependent regulation of OPC proliferation .. 75
ix
Chapter 1 Introduction
1
1.1 CNS cell diversity and oligodendrocyte
development
1.1.1 Development of CNS
The central nervous system (CNS) contains a diverse array of cellular types including neurons, macroglia (astrocytes, oligodendrocytes), and microglia. These cells need to be produced in a coordinated manner to ensure the proper formation and function of neural circuits1,2. Neurons and macroglia are derived from neuroepithelial progenitor (NEP) cells, and exposure to various morphogens informs them to generate subsets of neurons, astrocytes and oligodendrocytes. In particular, the generation of astrocytes and oligodendrocytes during development occurs through a step-wise process that includes the cessation of neurogenesis and the transition of progenitors towards glial fates. This is followed by a tightly controlled program of proliferation, migration, differentiation and maturation3. The GDE family of six-transmembrane proteins, play an important role in the temporal regulation of progenitor differentiation4.
GDE2 is expressed in post-mitotic neurons and its function utilizes enzymatic activity of its extracellular domain to induce motor neuron differentiation. GDE3, a close paralog of GDE2, is expressed in proliferating glial progenitor cells of the spinal cord, however its function in glial development has not been defined.
Here I show that GDE3 is a negative regulator of oligodendrocyte precursor cell
(OPC) proliferation and oligodendrocyte maturation in vivo and in vitro. RNAseq analysis of proliferating Gde3 KO OPCs implicates cytokine signaling as a misregulated pathway in this mutant. Gain-of-function and loss-of function studies identify
2 involvement of cytokine-related CNTF-CNTFRα signaling in GDE3-dependent control of
OPC proliferation. We report that GDE3 releases CNTFRα as a soluble factor via GPI- anchor cleavage and by a novel mechanism that involves recruitment of CNTFRα into an extracellular vesicle (EV) prior to its release from the EV membrane via cleavage of the GPI-anchor. This study uncovers a unique molecular mechanism utilized by GDE3 to regulate OPC development that may be applicable to other developmental processes.
1.1.2 Oligodendrocyte development
Within the CNS, oligodendrocytes provide trophic support to neurons and insulate axons for proper saltatory propagation of action potentials5. More recently, oligodendrocytes have also been discovered to make synaptic connections with neurons, suggesting additional contributions to neuronal plasticity and function6.
Oligodendrocyte development follows a series of sequential stages prior to maturation3.
These stages are characterized by changes in gene expression, protein profiles and oligodendrocyte morphology. At the onset of the neuron-glial switch, progenitor cells start expressing glial molecular markers that induce their glial fate7,8. Following specification, OPCs proliferate, migrate out of the ventricular zone and ultimately exit the cell-cycle to differentiate and myelinate axons. A sub-population of OPCs does not differentiate and exists within the CNS as quiescent cells termed adult OPCs9. In the spinal cord, the majority of OPCs are derived from the ventral domain, termed pMN, and are characterized by the expression of Olig2, NG2 and platelet-derived growth factor receptor alpha (PDGFRα). In addition to the pMN domain, a second wave of OPCs are
3 derived from dorsal progenitor domains (dP3–6) and constitute around 20% of all OPCs in the spinal cord3. OPCs retain their proliferative capacity as they migrate and populate the broader reaches of the spinal cord, and continue to require extracellular signals for proliferation and survival.
1.1.2.1 Oligodendrocyte Proliferation
Extrinsic factors are essential for proper development of OPCs, and can be broadly defined as factors that stimulate proliferation and inhibit differentiation, or factors that primarily drive the differentiation process. The quintessential factor for OPC proliferation and survival is platelet-derived growth factor A (PDGF-A), which signals through its cognate receptor PDGFRα. Overexpression of PDGF-A results in an elevated number of OPCs10; conversely, mice lacking PDGF-A show a severe decrease in OPCs and mature oligodendrocytes11. Furthermore, primary OPC cultures are normally maintained in medium containing PDGF-A, whose withdrawal induces their differentiation. Additional proliferation inducing mitogens include fibroblast growth factor
2 (FGF-2) and insulin-like growth factor 1 (IGF-1). FGF-2 promotes PDGFRα expression, amplifying the proliferative effects of PDGF-A on OPCs12. IGF-1 in combination with FGF-2 stimulates the ERK1/2 pathway to activate Cyclin D1; this induces the progression of the cell cycle through the G1 phase thus promoting OPC proliferation13. Cytokines are now recognized as having roles in oligodendrocyte proliferation during development and in response to CNS assault. For example, exogenous application of LIF in a demyelinating mouse model stimulates OPC proliferation14. Furthermore, in an experimental autoimmune encephalomyelitis mouse
4 model CNTF-null mice display a severe decrease in OPC proliferation following insult15.
The resident adult OPC population is essential for CNS recovery following injury. As such these cytokines have important roles in adult OPC proliferation, although their roles in the developing OPCs are not thoroughly understood. Our study uncovers the role of the CNTF-CNTFRα signaling axis to negatively regulate OPC proliferation. This function requires GDE3, thus establishing it as a novel regulator of OPC proliferation.
1.1.2.2 Oligodendrocyte Migration
OPCs migrate throughout the spinal cord and brain to achieve a uniform distribution, and their migratory routes are guided by external cues. PDGF-A and FGF-
2 work synergistically to stimulate OPC proliferation but act through independent mechanisms to instruct their migration16,17. Furthermore, chemokine CXCL12 acts as a chemoattractant18. Repulsive cues guiding OPCs include CXCR219 and several of
Semaphorin proteins (-3A, -4D, -4F)20–22. Together, these soluble and membrane bound motogenic factors establish spatial coordinates for OPCs to ensure their distribution prior to terminal differentiation.
1.1.2.3 Oligodendrocyte Differentiation
The differentiation of OPCs into mature oligodendrocytes coincides with changes in protein expression and in cellular morphology (Figure 1.1). OPCs undergoing maturation downregulate PDGFRα and NG2 as they transition from a bi-polar organization to complex multi-polar morphology. Further, maturing oligodendrocytes
5 co-express APC protein detected by CC-1 antibodies23, and 2’,3’-cyclic nucleotide 3’- phosphodiesterase (CNP) with subsequent expression of mature oligodendrocyte proteins such as Myelin-associated glycoprotein (MAG), myelin/oligodendrocyte glycoprotein (MOG), Oligodendrocyte glycoprotein (OMgp), Myelin basic protein (MBP),
Proteolipid protein (PLP) and alternatively spliced isoform DM20. Importantly, a combination of these markers is routinely used to assess the timing of oligodendrocyte maturation. OPC maturation and differentiation are regulated by a variety of soluble and membrane bound factors24. Negative regulators of differentiation include G-protein- couple receptor 17 (GPR17), which is expressed in late stage OPCs. GPR17 strongly inhibits OPC differentiation and maturation, and acts as an intrinsic signaling timer of oligodendrocyte differentiation and myelination25. Furthermore, glycosylphosphatidylinositol-anchored protein (GPI-AP) Contactin-1, has been reported to regulate OPC maturation and myelin formation26. Notch-1 signaling has differential effects on OPC differentiation depending on the bound ligand; axonal Jagged-1 and
Delta-1 inhibit OPC differentiation, while Contactin-1 (F3/contactin) promotes OPC differentiation27,28. Extrinsic factors also play important roles in regulating OPC differentiation. For example, withdrawal of proliferative factors such as PDGF-A promotes differentiation. In contrast, IGF-1 has unique roles in its ability to promote proliferation and also differentiation. Gain of function experiments show that IGF-1 administration causes robust increases in brain growth, the number of mature oligodendrocytes, and myelin gene expression, while IGF-1 null mice show decreases in mature oligodendrocytes29. Factors that negatively regulate oligodendrocyte proliferation and maturation have not yet been identified. Our study showcases GDE3’s
6 unique ability to inhibit OPC proliferation through CNTF-CNTFRα signaling and also suppress oligodendrocyte maturation.
Another component that drives oligodendrocyte maturation and myelination is neuronal activity. Optogenetic stimulation of neurons promotes OPC differentiation and increases myelination30. Taken together, these studies indicate that OPCs must integrate a great diversity of secreted and contact dependent signals to ensure their timely differentiation into myelinating oligodendrocytes and their subsequent myelination of axons.
1.1.3 Gp130 family of cytokines in CNS development
Our studies identified the cytokine signaling pathway as a potential mediator of
GDE3’s regulation of OPC proliferation. Cytokines are small (5-20kDa), pleotropic proteins initially identified as immune modulators involved in autocrine, paracrine and endocrine signaling31,32. Their functions are now recognized to extend far beyond the immune system to include the regulation of diverse functions in the central nervous system. Interleukin-6 (IL-6), Interleukin-11 (IL-11), Interleukin-27 (IL-27), leukemia inhibitory factor (LIF), oncostatin M (OSM), ciliary neurotrophic factor (CNTF), cardiotrophin-1 (CT-1), cardiotrophin-like cytokine (CLC), and neuropoietin (NP) all utilize transmembrane protein glycoprotein 130 (gp130) to implement cellular signal transduction33–35. Pleotropic signaling by these cytokines is attributed to additional receptors that associate with gp130. For example, LIF binds to a heterodimeric gp130/LIFR complex, while CNTF requires a tripartite receptor complex involving gp130/LIFR/CNTFRα for signal transduction (Table 1.1)35. Signaling downstream of
7 gp130 cytokines involves Janus Kinase/Signal Transducer and Activator of
Transcription (JAK/STAT) of which STAT3 is preferentially activated36 (Figure 1.2).
Additional signaling pathways also include mitogen activated protein kinase (MAPK) cascade, PI3K (phosphatidylinositol-3 kinase), mTOR (mammalian target of rapamycin) and AMPK (5’ adenosine monophosphate-activated protein kinase)37.
1.1.3.1 Gp130 signaling in neuroepithelial precursor cells
The gp130 family of cytokines has roles in the development of neuroepithelial precursor cells from which all neurons, astrocytes and oligodendrocytes are generated in the CNS. Loss of gp130 results in a reduced number of radial glial cells within the proliferative ventricular zone (VZ) in the forebrain, highlighting its role in regulating self- renewal during embryogenesis38. LIF and CNTF have also been shown to be important factors in the maintenance of progenitors within the VZ. Specifically, within the developing forebrain, CNTF/LIF/gp130 signaling maintains a subpopulation of distinct
VZ precursors that are required for the timely growth of early ventral forebrain38. In the same study, investigators found that CNTF-treated spinal cord explants show reduced proliferation of precursor cells within the VZ. In addition, astrocytes treated with LIF generate a more of a progenitor-like state capable of proliferation39. These observations suggest that gp130 may be integral to the maintenance of stem cell like populations in the nervous system.
8
1.1.3.2 Gp130 signaling in glia
Studies have identified additional roles for gp130 cytokines in gliogenesis.
Knockdown of gp130 in the embryonic cortex reduces the percentage of cells that express GFAP, suggesting roles for gp130 in astrogenesis40. However, deletion of LIF or CNTF in vivo has little effect on the generation of astrocytes indicating that other cytokines are involved. In support of this, CT-1 KO mice show decreased GFAP- expressing cells, similar to gp130 knockdown40. However, physiological roles for gp130 in astrogenesis remain unclear, because conditional deletion of gp130 in GFAP- expressing cells suggest it is not required for astrocyte formation41. This discrepancy may stem from the techniques employed in these studies. Cre-dependent ablation of gp130 in astrocytes may occur too late to address gp130 functions in astrogenesis, while caveats of the knockdown study include potential cytokine induction during the manipulation42. Further studies are required to clarify the role of gp130 signaling in astrogenesis. Injections of CNTF into chicken retina prior to toxic insult have been shown to reduce proliferation of Muller glia and give rise to progenitor-like cells43.
Meanwhile, intraocular injections of CNTF and FGF2 result in increased proliferation of
Muller glia44. These results highlight the dichotomy of cytokine signaling, and they reveal differential effects of CNTF that depend on the molecular context of other factors.
Roles for gp130 signaling pathways in oligodendrogenesis are emerging. LIF has been reported to regulate timing of oligodendrocyte development as is evident from in vivo studies of LIF KO mice, which show increased proliferation as assessed using
BrdU incorporation, and a concomitant increase in the number of PDGFRα and NG2+ cells45. LIF, CNTF and IL-11 have been shown to promote oligodendrocyte
9 differentiation and survival in vitro 46–49. The physiological roles of gp130 family cytokines in the regulation of oligodendrocyte remains to be fully understood. In the present study we identify the CNTFRα-gp130-LIFRβ receptor complex as a negative regulator of OPC proliferation. We find that addition of CNTF to OPC cultures inhibits their proliferation, while administration of neutralizing antibodies against gp130 in part abolishes CNTF’s effects. Importantly, we show that GDE3 mediates CNTF-dependent inhibition of OPC proliferation.
1.2 GDEs – Six-transmembrane domain GDPD
containing proteins
1.2.1 GDE2
GDE2 also known as GDPD5 is a six transmembrane protein with an extracellular enzymatic domain related to bacterial glycerophosphodiester phosphodiesterases (GDPD) (Figure 1.4A)4. GDE2 was discovered as a retinoic acid inducible gene that is expressed during spinal cord development50. GDE2 is expressed during the neurogenic period in the spinal cord and its enzymatic activity is required to induce motor neuron differentiation. GDE2 acts at the cell surface to cleave the GPI- anchor that tethers some proteins to the membrane, thereby regulating GPI-AP activity at the plasma membrane51. GDE2 induces motor neuron differentiation by cleaving and inactivating the GPI-AP ‘reversion-inducing cysteine-rich protein with kazal motifs’
(RECK)51. This results in disinhibition of the metalloprotease ADAM10, which removes the Notch ligand Dll1 from the surface of motor neurons. This series of events results in
10 downregulation of Notch signaling in adjacent pMN progenitors and induction of motor neuron differentiation. GDE2 activity is regulated through interaction with the thiol- reductase peroxiredoxin1 (Prdx1), which activates GDE2 by reducing an intramolecular disulfide bond between its N- and C- terminal domains52. GDE2 function extends beyond embryonic development. GDE2 null mice display degenerative neuropathology suggesting that GDE2 function is essential for neuron maintenance and survival53.
Other roles for GDE2 are emerging. GDE2 is also reported to guide pancreas development in zebrafish, and is found to play key roles in the differentiation of neuroblastoma through cleavage of the heparan sulfate proteoglycan GPC654,55.
1.2.2 GDE3
GDE3 (GDPD2) belongs to the GDE family of proteins and shares the same topology as GDE2, i.e. it is a six-transmembrane protein with an extracellular GDPD domain (Figure 1.4A). GDE3 plays central roles in controlling cellular proliferation and differentiation. Expression of GDE3 in cultured osteoblasts negatively regulates their proliferation56. Furthermore, xenografts of MDA-MB-231 cells overexpressing GDE3 leads to decreased tumor size compared to control57. Within the developing spinal cord
GDE3 expression is induced in the ventral progenitors at the onset of gliogenesis and encompasses the pMN domain from which OPCs arise (Figure 1.4B). Our lab has generated stable mouse lines that lack functional Gde3 and analysis of developing spinal cord showed that loss of GDE3 leads to delayed OPC specification and prolonged motorneuron differentiation (ChangHee Lee Thesis). These preliminary
11 studies suggest a role for GDE3 in nervous system development, and provide the foundation for my thesis work to study GDE3 function in gliogenesis.
1.3 Extracellular vesicles
Several release modalities of GPI-APs have been identified58. In addition to lipolytic cleavage of the GPI moiety, as exhibited by the GDE family members, GPI-APs can be released with their anchor intact and tethered to the membrane of an extracellular vesicle (EV)58. EVs are composed of a lipid bilayer containing membrane associated proteins enclosing soluble hydrophilic cytosolic factors that include proteins and RNA. Most if not all cell types release EVs and as such they can be found in body fluids such as blood, urine, serum and cerebral spinal fluid59. Within the CNS EVs have been reported to be released from neurons, astrocytes, oligodendrocytes and microglia, and have been proposed to function as mediators of intercellular communication60. In addition to their role in cellular signaling they are also proposed to propagate disease through transfer of proteins such as prion-like proteins and Tau61,62. EVs are a heterogenous group of different sized vesicles (50nm to 1000nm) with varying contents of proteins, lipids and RNA. EV nomenclature suggests the use of “exosomes” for particles less than 100nm in diameter and “microvesicles,” or “ectosomes,” for particles over 100nm63. However, there are currently no markers to distinguish differentially sized vesicles and as such the broadly encompassing term of EVs will be used.
Identification of mechanisms in EV biogenesis, release, and signaling are critical for understanding their roles in developmental and pathological processes.
12
1.3.1 Biogenesis of EV
EVs can be released from the cell through two pathways – 1) Multivesicular bodies (MVB) fusing with the plasma membrane or 2) outward budding and fission of the plasma membrane64. Most studies to date have focused on the production of
MVBs. MVBs are generated from the inward budding of vesicles from endosomes creating intraluminal vesicles (ILVs). Upon fusion of MVBs with the plasma membrane,
ILVs are released into the extracellular space as EVs. There are multiple machineries involved in the generation of ILVs within the MVB (Figure 1.3). Formation of ILVs can be accomplished through tetraspanins (CD9 and CD63), endosomal sorting complexes required for transport (ESCRTs), and specific lipid species (ceramides)65. These pathways can act in concert, or independently of each other depending on the cellular context. Biogenesis of EVs can be broadly categorized into two molecular mechanisms
1) ESCRT-dependent and 2) ESCRT-independent pathways65–67. ESCRTs are of particular interest as they govern selection of cargo and they are drivers of intraluminal membrane budding due to their ability to undergo physical membrane-remodeling and scission68,69. As such, certain components of the ESCRT machinery can be found in the EVs such as ALIX and TSG101. ESCRT independent pathways include the syndecan-syntenin-ALIX release pathway which is regulated by ADP ribosylation factor
6 (ARF6), phospholipase D2 (PLD2), and syntenin70. Studies in immortalized oligodendrocyte cultured cells have identified further contributions of the ceramide pathway, where ceramide generation through the neutral sphingomyelinase (nSMase) enzyme can induce ILV invagination that recruits specific cargo of PLP, CD63 and
TSG10171. Taken together, there are at several pathways through which EVs are
13 generated and these pathways can recruit distinct cargo for release and so can contribute to EV heterogeneity.
1.3.2 Release mechanisms of EV
Once MVBs with the ILVs haves been generated, transport to the plasma membrane and fusion is required for EV release. This process is regulated by multiple factors showing specificity for release of certain cargo in the EVs. RABs have been implicated in facilitating fusion of MVB with plasma membrane, and these include
RAB11 and RAB35, RAB7, RAB27A/B among others72. RAB proteins are involved in intracellular vesicle transport between different cellular compartments. Loss of distinct
RABs has been found to result in decreased release of certain EV cargo. For example, silencing Rab27 or its effectors Slp4 and Slac2b in HeLa cells decreases EV secretion with concomitant decreased release of CD63, ALIX, and TSG101 proteins73. Loss of
RAB7 in the MCF-7 breast cancer cell line results in loss of EVs containing syntenin70, however knockdown of RAB7 in HeLa cells revealed no significant changes related to
EV release73. These results suggest RAB proteins are important components for EV release, but their requirement differs based upon cargo and cell type.
1.3.3 EV content and signaling
EVs contain various proteins, lipids and nucleic acids65. All these components can potentially contribute to EV-dependent signaling. Studies have shown that EVs can be internalized by acceptor cells via direct fusion with plasma membrane or
14 endocytosis74,75. This would result in the deposition of soluble EV cargo into the cytosol of acceptor cells integration of and membrane-bound components in the plasma membrane. EVs are enriched for distinct lipid species including cholesterol, diacylglycerol, sphingomyelin and ceramides among others76. These lipid species are components of detergent-resistant domains found on the plasma membrane of cells, which are also enriched in GPI-APs. Indeed, GPI-APs CD55 and CD59 have been found to be released in EVs from erythrocytes and this release is speculated to modulate their senesence58. In addition to GPI-APs, EVs can transport transcription factors, signaling proteins, mRNA and miRNA. Analysis of retinal progenitor cells shows that their EVs contain mRNA, miRNA and proteins associated with multipotency and that this cargo is taken up by target retinal progenitor cells77. Current studies are looking into factors that facilitate the uptake of EVs. Among studied these factors are heparan sulfate proteoglycans (HSPGs). Blocking these proteins on recipient cells with heparin results in reduced uptake of EVs78,79. In addition to uptake, EVs can signal between cells via direct contact with plasma membrane receptors of recipient cells.
EVs from neural stem cells transfer IFN-gamma through Ifngr1 to activate STAT1 signaling in target cells80. This study further shows that IFN-gamma bound to Ifngr1 on
EVs has much higher stability, ~100-fold more, compared to its soluble counterpart.
These results suggest that EVs may constitute a stable reservoir of signaling molecules.
Of note, GPI-APs harboring complete GPI anchors are capable of transfer from donor to acceptor cells in vitro and in vivo81,82 . Therefore, EV-associated GPI-APs could be transferred to distant acceptor cells. No mechanism exists for selective GPI-AP
15 recruitment into EVs. Our present study identifies GDE3 as being capable of selective recruitment of specific GPI-APs for EV release.
1.4 Summary and Specific Aims
Our laboratory has identified the novel GDE family of six-transmembrane proteins, each of which contains an extracellular GDPD domain, as mediators of cell differentiation during CNS development. GDE2, has been shown to play a pivotal role in regulating the timing of neurogenesis and motor neuron differentiation. GDE2 function is dependent on the catalytic activity of its extracellular GDPD domain, which is capable of cleaving and releasing GPI-APs from the cell surface. GDE3, another member of the GDE family, is induced in progenitor cells undergoing the neuron-glial switch and its expression is maintained in astroglial precursor cells, OPCs and differentiated astrocytes. Like GDE2, GDE3 is capable of GPI-anchor cleavage when co-expressed with certain GPI-APs in HEK293 cells. Based on these observations, I hypothesize that GDE3 regulates glial specification and differentiation through release of target GPI-APs. The goals of my thesis project are to define the requirement for
GDE3 in glial cell development, and to determine the mechanisms involved.
Aim 1: To define the requirement of GDE3 in glial proliferation and differentiation
Aim 2: To determine if GDE3 function requires extracellular GPI-anchor cleavage activity
Aim 3: To determine the downstream pathways regulated by GDE3 in glial proliferation 16
Figure 1.1 Markers of oligodendrocyt e lineage cells through specification and maturation. The schematic illustrates changes in protein expression and morphology of oligodendrocyte lineage cells. OPCs express Olig2, PDGFR α and NG2. Olig2 is expressed throughout the course of oligodendrocyte development. Maturation of oligodendrocyte can be identified through loss of expression of PDGFR α and NG2 and onset of mature oligodendrocyte proteins such as CC1, CNP, MAG, PLP and MBP (in red). Adapted from Silberis et al. 2010 83
17
Figure 1 .2 Signaling pathways implicated in CNTFR α signaling Th e signaling pathways activated by CNTF. CNTFR α binds CNTF and results in heterodimerization of gp130 and LIFR. Soluble versions of CNTFR α are biologically active and capable of activating gp130 -LIFR signaling. Adapted from Pasquin et al.
2015 84
18
Figure 1.3 Molecular machineries of extracellular vesicle biogenesis. Multiple machineries are involved in biogenesis of intralumin al vesicles of multivesicular bodies EVs. Endosomal sorting complex required for transport (ESCRT) components, lipids, and tetraspanins have been described, but whether each acts in different MVBs, or if they can simultaneously act on the same MVB, is not known. Adapted from
Colombo et al. 2014 65
19
Figure 1. 4 GDE topology and expression of GDE3 in the developing spinal cord. (A) Topology of GDE fam ily of proteins. N - and C - termini face the cytoplasm.
Enzymatic GDPD domain is located in the extracellular space. (B) Schematic of mRNA expression of Gde3 within developing spinal cord at the onset of gliogenesis
(embryonic day 12.5). Pax7 demarks th e dorsal progenitor domains. Gde3 expression is restricted to the ventral progenitor domains and encompasses the pMN domain from which OPCs are generated (Adapted from ChangHee Lee Thesis)
20
Table 1.1 gp130 Cytokines and their receptor components.
Adapted from (White and Stephens, 2011)
21
Chapter 2 GDE3 regulates OPC proliferation and differentiation in vivo
22
2.1 Introduction
Previous studies showed that Gde3 is expressed within the ventricular zone of the developing spinal cord, encompassing the pMN domain from which motor neurons and subsequently oligodendrocytes are generated. GDE3 expression initiates within the progenitor cells undergoing the neuron-glial switch and is maintained in OPCs, astrocyte precursor cells (APCs) and astrocytes, implicating GDE3 in glial development.
Furthermore, studies of Gde3 in osteoblast cells suggested that GDE3 negatively regulates cellular proliferation56. Given its expression pattern and role in controlling cellular proliferation, we hypothesize that Gde3 regulates the proliferation of glial progenitors in the spinal cord. To test this hypothesis, we explored GDE3 regulation of ventral glial progenitor cell cycle kinetics in the spinal cord and assessed the consequences of Gde3 loss on glial cell development.
2.2 Results
2.2.1 GDE3 regulates cell cycle kinetics of ventral progenitors in spinal
cord
To determine if GDE3 regulates the proliferation of ventral progenitors, we measured the length of S-phase and total cell-cycle length in Gde3 WT and KO progenitors at E12.5. We defined ventral progenitors by their expression of Nkx6.1 which encompasses progenitor domains where Gde3 is expressed (Figure 1.4B). We utilized a previously described protocol that involves sequential injection of the thymidine analogues BrdU and IdU at an interval of 1.5 hours prior to harvesting
23 embryos85. Calculation of the number of cells that have incorporated BrdU and IdU, or just BrdU, in conjunction with the known time interval between injections allows for calculation of the length of S-phase and total cell cycle length of proliferating cells within a specified region85. Utilizing this approach, we found that Gde3 KO ventral progenitors have a shorter S-phase at 8.50 ± 0.52 hours compared to 11.61 ± 0.69 observed in WT littermates (Figure 2.1C). Furthermore, Gde3 KO ventral progenitors possess decreased cell cycle length of 20.97 ± 2.98 hours compared to 27.20 ± 2.27 hours WT animals (Figure 2.1D). These results suggest that GDE3 negatively regulates the proliferation dynamics of ventral progenitors.
2.2.2 Loss of Gde3 results in increased OPC production and differentiation
in vivo
Our results described in the previous section demonstrate that loss of Gde3 results in increased proliferation of ventral progenitors that will give rise to glial precursors, including OPCs and APCs. To examine the consequences of these proliferation perturbations on glial development, we focused our studies on OPC formation. OPCs are derived from the pMN domain located within the ventral progenitor domains that express GDE3 (Figure 1.4B)3. Once specified, Olig2+ OPCs migrate outside of the pMN domain to populate the spinal cord. Because the loss of Gde3 results in increased proliferation of ventral progenitors, we expect the number of Olig2+
OPCs to increase when GDE3 function is disrupted. To test this hypothesis, we utilized a genetic mosaic model to directly examine the behavior Gde3 KO OPCs relative to WT
OPCs. Gde3 is expressed from the X-chromosome. Female mammals possess two X-
24 chromosomes, with one of them undergoing X-chromosome inactivation, and therefore
Gde3+/- animals are natural mosaics with each cell expressing one allele from the X- chromosome. We used a HprtGFP mouse that contains an X-chromosome reporter which expresses GFP, and which will co-segregate with a WT Gde3 allele (Figure
2.2A)86. We generated mosaic female embryos that were either HprtGFP;Gde3+/+ or
HprtGFP;Gde3+/- by breeding HprtGFP;Gde3+/y with a heterozygous Gde3 female –
Gde3+/-. In the HprtGFP;Gde3+/+ animals, all cells are wildtype; however, in
HprtGFP;Gde3+/- animals, cells that are GFP+ are WT, while GFP- cells are Gde3 KO
(Figure 2.2B-D). Analysis of spinal cords from E14.5 mice showed that GFP- Gde3 KO cells constitute a higher proportion of the Olig2+ population in HprtGFP;Gde3+/- animals when compared to GFP- Gde3 WT cells in HprtGFP;Gde3+/+ littermates, (67% vs. 59% respectively Figure 2.2E). These results support our previous findings illustrating that loss of GDE3 leads to increased proliferation of ventral progenitors. If the loss of GDE3 results in a hyperproliferation phenotype, then we surmise that the proportion of Olig2+ cells composed from Gde3 KO would be even higher at later timepoints. Analysis of cellular contribution to Olig2+ cells at P6 was surprising – GFP- Gde3 KO cells now contributed significantly less to the Olig2+ population compared to their GFP- Gde3 WT counterparts, 40% vs. 56% respectively (Figure 2.2E). No significant changes in cell death were observed, as assessed by cleaved caspase3 staining between
HprtGFP;Gde3+/- and HprtGFP;Gde3+/+ animals (Figure 2.3C). One possible explanation for the reduced contribution of Gde3 KO cells to the Olig2+ population at P6 is that
Gde3 KO OPCs might differentiate ahead of schedule. If so, Gde3 KO OPCs might exit the cell cycle earlier, and allow the WT OPCs to proliferate and contribute more
25 extensively to the overall Olig2+ population at P6. To test for this possibility, we examined the proportion of mature oligodendrocytes identified by their expression of both CC1 and Olig2 in P14 animals (Figure 2.4A). In HprtGFP;Gde3+/- animals, 68% of
GFP-Olig2+ Gde3 KO cells express CC1, while only 47% of GFP+Olig2+ Gde3 WT cells expressed CC1 (Figure 2.4B). Within the control HprtGFP;Gde3+/+ control animals there were no significant differences between GFP+ and GFP- CC1+Olig2+ cell populations. This suggests that loss of GDE3 results in increased propensity of OPCs to differentiate into mature oligodendrocytes. Although the proportion of Gde3 KO
Olig2+ cells at P6 is significantly reduced, it is possible that Gde3 KO cells retain increased proliferative potential compared to WT Olig2+ precursors. Our analysis of the proliferative marker Ki67 shows GFP-Olig2+ Gde3 KO cells make up a significantly larger proportion of proliferating OPCs compared to GFP-Olig2+ WT cells at P6 (63% vs. 48%, respectively, Figure 2.5A). Consistent with this notion, by P14 GFP-Olig2+
Gde3 KO cells rebound to make up equivalent proportions of Olig2+ cells as their GFP-
WT counterparts (Figure 2.2E).
2.3 Summary
Our genetic studies suggest that GDE3 negatively regulates progenitor cell proliferation in the developing spinal cord. Indeed, loss of GDE3 affects cell-cycle kinetics of glial progenitors, resulting in increased proliferation and a shortened cell cycle. We utilized the fact that Gde3 is located on the X-chromosome to generate a genetic mosaic mouse model to assess the consequences of Gde3 loss on OPCs.
GDE3 loss results in increased generation of OPCs and their differentiation into mature
26 oligodendrocytes. Furthermore, given the nature of the mosaic model where both WT and KO cells intermix, our studies suggest that GDE3 acts cell autonomously to regulate OPCs proliferation and differentiation. Taken together, our results suggest that
GDE3 functions to control the pace of OPC development.
27
Figure 2.1 Deletion of Gde3 increases proliferation of VZ progenitors (A and B) Images of the ventral VZ progenitors in E12.5 spinal cord. Arrowheads highlight cells that incorporated IdU. (C and D) At E12.5, BrdU/IdU ba sed measurements reveal decreased S -phase length (p* = 0.007, n = 4) and a shortening of the total cell cycle (p* = 0.016, n = 8) within ventral VZ progenitors.
28
29
Figure 2.2 Gde3 null OPCs exhibit heightened proliferation (A) Breeding scheme for mosaic analysis of OPC proliferation. The HprtGFP reporter
+/y gene cotransmits with a WT Gde3 allele, thus crossing an HprtGFP;Gde3 male to a
+/- +/+ Gde3 female yields two possible female offspring genotypes. HprtGFP;Gde3 females will have GFP+ and GFP- cells that are both Gde3 WT, while
+/- HprtGFP;Gde3 females will have GPF+ cells that are Gde3 WT and GFP- cells that
+/- are Gde3 KO. (B-D) Example image from HprtGFP;Gde3 spinal cord at E12.5.
Arrow highlights a Gde3 WT OPC and arrowhead labels a Gde3 KO OPC. (E)
Quantification of the Gde3-HprtGFP mosaic analysis. Gde3 KO GFP- cells in
+/- HprtGFP;Gde3 animals at E14.5 comprise a larger proportion of the total Olig2
+/+ population than Gde3 WT GFP- cells in HprtGFP;Gde3 controls (p* = 0.012, n = 4
WT, 5 Het). The inverse relationship is observed at P6 (p* = 0.013, n = 3), and no proportional differences are detected at P14.
30
Figure 2.3 Gde3 regulates OPC proliferation and differentiation without changes in cell death.
+/- (A and B) Image of P6 HprtGFP;Gde3 spinal cord stained for Olig2 and the cell death marker cleaved caspase-3. Arrow indicates a double positive cell. (C)
Quantification of cleaved caspase-3 reveals no differences between the Olig2
+/+ +/- populations in HprtGFP;Gde3 versus HprtGFP;Gde3 animals at P6 (p = 0.229 , n
= 3).
31
B
A
Figure 2.4 Gde3 regulates maturation of oligodendrocytes
+/- (A) Representative image of a P14 spinal cord from an HprtGFP;Gde3 animal stained with CC1. Arrowheads denote GFP- Gde3 KO OLs and arrows highlight GFP+ Gde3
WT OLs. (B) Gde3 KO GFP- cells constitute a larger proportion of the total CC1+ Olig2
+/- +/+ population in HprtGFP;Gde3 than Gde3 WT GFP- cells in HprtGFP;Gde3 control mice (p* = 0.005 , n = 3 WT, 6 Het).
32
Figure 2. 5 Gde3 null oligodendrocytes display increased proliferation (A) Quantification of actively proliferating (Ki67+) oligodendrocytes (Olig2+) in a P6 spinal cord. Gde3 KO GFP - cells compose a larger proportion of Ki67+Olig2+
+/ - +/+ population in HprtGFP ;Gde3 than Gde3 WT GFP - cells in HprtGFP ;Gde3 control mice (p* = 0 .013 , n = 3).
33
Chapter 3 GDE3 regulates proliferation and differentiation of purified OPCs
34
3.1 Introduction
Our analysis of Gde3 KO mice suggests that loss of GDE3 results in increased
OPC proliferation and differentiation. To enhance our understanding of GDE3’s function in oligodendrocyte development, we isolated OPCs from mouse spinal cord and cortices to examine the proliferation and differentiation capacity of Gde3 null OPCs.
Furthermore, this in vitro model system gives us the ability to determine if GDE3 is sufficient to inhibit OPC proliferation and to test cell-autonomy of GDE3 function.
3.2 Results
3.2.1 GDE3 is necessary to inhibit OPC proliferation
To gain greater insight into GDE3’s ability to regulate OPC proliferation, we have examined cultured OPCs purified from P0 spinal cords or P2-P4 cortices of Gde3 KO and WT littermates. The primary OPC cultures were maintained in chemically defined proliferative media containing growth factors PDGF and CNTF. To assess OPC’s proliferative capacity, cultures were pulsed for 2 hours with IdU to label cells in S-Phase and stained with NG2 to confirm OPC identity. Proliferating OPCs were defined as the percent of NG2 cells that had incorporated IdU. Cultures prepared from spinal cord showed that compared to WT littermates, loss of Gde3 results in an approximately 35% increase in proliferating OPCs (Figure 3.1E). This is in line with our in vivo phenotypes, where we observe increased generation of Olig2+ cells in addition to shortened cell cycle length in Gde3 KO animals (Chapter 2). Furthermore, purified cortical Gde3 KO
OPCs display an analogous increase of 27% in IdU incorporation compared to WT
35 littermates (Figure 3.1E). Taken together, these observations reveal shared functionality of GDE3 in modulating spinal and cortical OPC proliferation.
3.2.2 GDE3 controls the maturation of OPCs
To examine the consequences of GDE3 loss to OPC maturation in an in vitro setting, we withdrew the proliferative growth factor, PDGF, from the OPC media to initiate differentiation of OPCs into mature oligodendrocytes. Mature oligodendrocytes can be defined by the expression of the myelin protein MBP. We analyzed the ratio of mature oligodendrocyte MBP+ cells between WT and Gde3 KO cortical cultures. Three days following PDGF withdrawal Gde3 KO cultures show an increased ratio of mature oligodendrocytes expressing MBP compared to their WT counterparts, 16% and 6%, respectively (Figure 3.2C). This observation suggests that the loss of GDE3 results in accelerated maturation of OPCs in vitro and is in line with our studies in the mosaic
GDE3 model displaying accelerated differentiation.
3.2.3 GDE3 is sufficient to inhibit OPC proliferation
Loss of GDE3 results in increased proliferative capacity of both spinal and cortical OPCs cultured in vitro. To test if GDE3 expression is sufficient to inhibit OPC proliferation, we performed a gain of function experiment by transfecting constructs expressing GDE3 into OPCs purified from WT cortices. Because of the similar proliferation phenotype between spinal cord and cortical OPCs grown in vitro, we performed transfections on cortical OPCs since large numbers of cortical OPCs are more easily isolated. GDE3 was subcloned into a bicistronic construct that contains an
36
IRES sequence downstream of GDE3 that expresses GFP; this strategy facilitates the identification of transfected OPCs by GFP expression (Figure 3.3A). The percentage of transfected OPCs (NG2+GFP+) that had incorporated IdU (IdU+NG2+GFP+) was quantified. Compared to expression of GFP alone, expression of GDE3 resulted in a
20% reduction of OPCs incorporating IdU (Figure 3.3B). This observation suggests that the expression of GDE3 is sufficient to negatively regulate the pace of OPC proliferation. To examine the autonomy of GDE3 function, we quantified the proliferation capacity of non-transfected OPCs (NG2+GFP-). Strikingly, the proliferation of non-transfected OPCs in conditions of GFP or GDE3 transfection was found to be equivalent (Figure 3.3B). These observations suggest that GDE3 inhibition of OPC proliferation is cell-autonomous and does not affect the proliferation of neighboring untransfected OPCs.
3.3 Summary
To determine the role of GDE3 in OPC proliferation and differentiation more precisely, we isolated and cultured primary OPCs from WT and Gde3 KO animals. We find that cultured Gde3 KO OPCs display heightened proliferation and accelerated maturation. Furthermore, gain of function experiments with cultured OPCs showed that expression of GDE3 is sufficient to inhibit OPC proliferation, and that GDE3 acts in a cell autonomous manner in this capacity. The in vitro OPC culture system recapitulates the in vivo proliferation and differentiation Gde3 KO phenotypes and accordingly is a suitable model for further inquiry into GDE3 function.
37
Figure 3.1 Spinal cord and cortical OPCs lacking Gde3 display heightened proliferation
(A-D) Images of cultured OPCs isolated from neonatal spinal cord (A and B) or early postnatal brain (C and D). Arrows highlight examples of proliferating OPCs (IdU+NG2+) in each condition. (E) Graph quantifying the heightened degree of IdU incorporation in
Gde3 KO cultures derived from spinal cord (p* <0.001; n = 6 WT, 7 KO) and cortex (p*
= 0.002; n = 9).
38
Figure 3 .2 Gde3 null OPCs display increased maturation in vitro (A and B) Images of cultured OPCs isolated from early post -natal cortex induced to differentiate through withdrawal of PDGF for 3 days. Arrowheads highlight examples of mature oli godendrocytes (MBP+) in each condition. (C) Graph quantifying the heightened degree of maturation of oligodendrocytes in Gde3 KO cultures (p* = 0.03; n = 3)
39
Figure 3.3 Expression of GDE3 inhibits OPC proliferation (A-A’’’) Example image from WT OPC cultures transfected with GDE3 and labeled with
IdU. Arrows show transfected IdU+ OPCs, and arrowheads denote transfected IdU-
OPCs. (B) Quantification of IdU incorporation following GFP versus GDE3 transfection.
Among transfected cells (GFP+), GDE3 significantly reduces OPC proliferation (p* <
0.001, n = 12). Among non-transfected cells (GFP-) in the same well, IdU incorporation is unchanged between conditions (p = 0.482, n = 12).
40
Chapter 4 Identifying pathways that mediate GDE3-dependent regulation of OPC proliferation
41
4.1 Introduction
Our analysis of GDE3 function suggests that GDE3 regulates the pace of oligodendrocyte development by negatively controlling OPC proliferation and differentiation. The critical questions that remain are: how does GDE3 exert its influence, and specifically what molecular target(s) does GDE3 act on? To address these questions, we performed RNA-sequencing on isolated WT and Gde3 KO OPCs to evaluate global transcriptional changes elicited by the disruption of GDE3 function. We performed functional enrichment analysis on the differentially expressed genes and successfully identified misregulated pathways in Gde3 KO OPCs. By combining this approach with known mechanisms of GDE3 function, we identified the CNTF pathway and the GPI-anchored protein CNTFRα as putative mediators of GDE3-dependent inhibition of OPC proliferation.
4.2 Results
4.2.1 RNA-seq approach
OPC proliferation precedes differentiation, so we decided to first elucidate pathways regulated by GDE3 in OPC proliferation. To determine the pathways that
GDE3 utilizes to regulate OPC proliferation, we performed RNAseq analysis using cDNAs generated from OPCs isolated from three biological replicates of P0.5 WT and
Gde3 KO spinal cords. Purified OPCs were grown in proliferative media for 16 hours prior to mRNA isolation and library construction. The quality of the library was verified using BioAnalyzer prior to sequencing. 50 bp paired end reads were generated on an
42
Illumina HiSeq 2500 system yielding 50-61 million reads per sample. Mapping rates to the mouse genome were 98%, with 86% representing exonic reads, indicating that the sequences obtained were of high-quality. To identify transcripts that were misregulated in Gde3 KO OPC samples, we focused on genes that met the criteria of differentially expressed to a statistically significant level using cutoff points of p < 0.002, and q <
0.05. Comparison of WT and Gde3 KO sequences identified a total of 96 differentially regulated genes, of which 53 genes are upregulated and 43 genes are downregulated in Gde3 KO OPCs (Figure 4.1A).
4.2.2 Functional annotation of differentially enriched genes
Genes have ascribed biological, molecular and cellular identifiers, and this is termed “gene ontology87”. Analysis of gene ontology descriptors of differentially expressed gene sets can identify overrepresented pathways within the gene set.
Functional enrichment analysis was performed for the 96 differentially regulated genes derived from RNAseq analysis using the STRING database (v10.5). This analysis identified multiple pathways perturbed in Gde3 KO OPCs compared with WT OPCs. In support of GDE3 function in OPC proliferation and differentiation, genes encoding proteins involved in regulation of cell differentiation (GO.0045595) and regulation of developmental processes (GO.0050793) were identified (Figure 4.1B). Notably, we detected significant changes in pathways involving cellular response to cytokines
(GO.0034097) and extracellular exosomes (GO.0070062), highlighting these pathways as potential mediators of GDE3-dependent regulation of OPC proliferation.
43
4.2.3 CNTFRα, a candidate GPI-AP mediating cytokine signaling
Cytokine signaling plays a central role in the development of the central nervous system, including roles in cellular proliferation42 (Chapter 1.1.3). We thus examined the possibility that the alterations in cytokine pathway regulation contribute to the increased proliferative rates of Gde3 KO OPCs. Prior work from our laboratory and others shows that GDE3 can regulate the activities of GPI-APs by releasing them from the cell surface into the extracellular space. Accordingly, we hypothesized that GDE3 regulates OPC proliferation through release of a GPI-anchored protein that mediates cytokine signaling.
With this in mind, we curated a database of annotated GPI-APs in the mouse genome, identified candidates with known roles in cytokine signaling and cross- referenced them to published expression datasets of genes that are enriched in
OPCs88. We identified six GPI-APs that are involved in cytokine signaling, two of which are abundantly expressed in OPCs: CNTFRα and Sema7a (Table 4.1). Strikingly,
CNTFRα is implicated in OPC proliferation, maturation, differentiation, and survival46,47,89. Furthermore, CNTFRα is the ligand binding component of a multiprotein receptor for CNTF, one of the factors present in the OPC proliferative media90. Western blot analysis of cultured OPCs shows that GDE3 and CNTFRα are coexpressed in
OPCs (Figure 4.2). To test if CNTFRα is a physiological substrate of GDE3, we subjected dissected cortex to detergent fractionation using Triton X-114. Soluble proteins and cleaved, and soluble CNTFRα should localize to detergent poor fraction.
As a positive control, we recover soluble cytosolic protein GAPDH in the soluble fraction confirming successful Triton X-114 partitioning (Figure 4.3A). We quantified the intensity of the CNTFRα band as a ratio of soluble CNTFRα to total CNTFRα in the
44 input. We detect a reduction of approximately 25% in the generation of soluble
CNTFRα in GDE3 KO mice compared to their WT counterparts (Figure 4.3B). Based on these observations, CNTF signaling and CNTFRα constitute strong candidates for mediating GDE3-dependent effects on OPC proliferation.
4.3 Summary
Our analysis of RNAseq data obtained from WT and Gde3 KO OPCs suggests the possibility that cytokine signaling pathways mediate GDE3-dependent inhibition of
OPC proliferation. By examining cytokines that are known to include GPI-APs in their signal transduction pathway, we identified the CNTF pathway as a promising candidate since the ligand-binding component of the CNTF receptor is the GPI-anchored protein
CNTFRα.
45
Figure 4.1 RNAseq results from Gde3 KO OPCs (A) Volcano plot visualizing annotated genes identified by RNA sequencing from cDNA libraries generated from proliferating spinal cord WT and Gde3 KO OPCs. 53 genes were significantly increased (red), 43 genes were significantly decreased
(blue) and the 16433 remaining transcripts showing no difference between WT and
Gde3 KO (grey). Significance cutoff for differential gene expression was q ≤ 0.05 (p ≤
0.002) (B) Functional annotation analysis of the genes differentially expressed in the absence of Gde3 have significant enrichment for proteins involved in the regulation of developmental process (p* = 0.001), the regulation of cell differentiation (p* = 0.03), the response to cytokine (p* = 0.001) and extracellular exosomes (p* = 0.008).
Constituent genes for each network are plotted amongst all mapped RNAseq genes in the volcano plot along with the FDR-corrected significance values for each GO term.
46
Figure 4. 2 OPCs express GDE3 and CNTFRα Surface biotinylation of WT and Gde3 -/y OPC cultures. Western blots show that GDE3 and CNTFRα are expressed on the surface of OPCs. Na -K-ATPase is used as a posi tive control.
47
Figure 4. 3 Levels of soluble CNTFRα are decreased in Gde3 KO animals (A) Western blots of early post -natal (P2 -P4) cortex of total lysates (Input) and following
Triton X -114 partitioning. Soluble prote in GAPDH is recovered in the soluble fraction
(T2). (B) Quantification of the ratio of soluble CNTFRα to the total CNTFRα in the input.
Gde3 KO shows decreased levels of soluble CNTFRα compared to WT (*p = 0.013, n =
11 WT; 4 KO)
48
Table 4.1 GPI-APs with cytokine signaling
(A) List of GPI-anchored proteins with known roles in cytokine signaling (Uniprot).
CNTFRα has the highest expression in OPCs and its ligand CNTF is a known regulator of OPC activity.
49
Table 4.2: RNA-sequencing genes that were significantly altered in Gde3 KO OPCs Number of Affected genes: 96
Gene ID Gene Locus Gde3 KO Gde3 WT Fold Change (log2) p value q value XLOC_032941 Mitf chr6:97807051-98021349 26.8648 1.36012 4.30391125 5.00E-05 0.0156975 XLOC_004033 Tmcc3 chr10:94311948-94612084 120.852 24.1674 2.322107172 5.00E-05 0.0156975 XLOC_029122 Isg15 chr4:156199423-156200818 7.93802 1.34037 2.566147895 5.00E-05 0.0156975 XLOC_018882 Ifit3 chr19:34583530-34588731 2.70457 0.304181 3.152397292 5.00E-05 0.0156975 XLOC_028863 C1qa chr4:136895916-136898803 21.4913 7.36083 1.545812396 5.00E-05 0.0156975 XLOC_028861 C1qb chr4:136880128-136886187 25.3362 9.35195 1.437861039 5.00E-05 0.0156975 XLOC_033069 Usp18 chr6:121245905-121270916 2.48346 0.488183 2.346857548 5.00E-05 0.0156975 XLOC_014793 Rtp4 chr16:23609918-23614222 2.72191 0.571574 2.251607168 5.00E-05 0.0156975 XLOC_007962 Lgals3bp chr11:118392749-118403903 9.33925 2.76843 1.754238654 5.00E-05 0.0156975 XLOC_004908 Lyz2 chr10:117277333-117282274 135.788 76.3509 0.830638922 5.00E-05 0.0156975 XLOC_028862 C1qc chr4:136889803-136893065 25.0492 9.2053 1.444227884 5.00E-05 0.0156975 XLOC_039284 Bst2 chr8:71534263-71537437 14.2063 5.44139 1.384483717 5.00E-05 0.0156975 XLOC_030248 Oasl2 chr5:114896935-114912239 2.86762 0.569796 2.33133646 5.00E-05 0.0156975 XLOC_038460 Lpl chr8:68880479-68907448 21.4016 10.2562 1.061222357 5.00E-05 0.0156975 XLOC_024672 Gbp3 chr3:142559901-142573209 2.65078 0.609018 2.121860166 5.00E-05 0.0156975 XLOC_016526 Adgre1 chr17:57358685-57483529 9.54143 4.78497 0.995695623 5.00E-05 0.0156975 XLOC_013571 Cyth4 chr15:78597046-78622019 2.4248 0.879892 1.462467397 5.00E-05 0.0156975 XLOC_022757 Siglec1 chr2:131068114-131086765 1.20221 0.426985 1.493431632 5.00E-05 0.0156975 XLOC_014891 Hcls1 chr16:36934982-36963212 6.79269 3.10363 1.130026438 5.00E-05 0.0156975 XLOC_030335 Myl2 chr5:122100950-122140823 0.251732 0.0407529 2.626913991 5.00E-05 0.0156975 XLOC_012075 Cpne6 chr14:55510444-55517440 0.547495 1.96818 -1.845944474 5.00E-05 0.0156975 XLOC_011651 Prrxl1 chr14:32599406-32649246 0.191034 1.12338 -2.555944688 5.00E-05 0.0156975 XLOC_023454 Stmn2 chr3:8509342-8561606 20.0226 39.0127 -0.962314523 5.00E-05 0.0156975 XLOC_006608 Nefh chr11:4938753-4948064 0.41425 1.61223 -1.96048397 5.00E-05 0.0156975 XLOC_004607 Susd2 chr10:75636618-75644008 0.33774 1.53914 -2.188139506 5.00E-05 0.0156975 XLOC_015921 Lix1 chr17:17402671-17459387 0.689718 2.08438 -1.595539794 5.00E-05 0.0156975 XLOC_007559 Ngfr chr11:95568817-95587735 3.88879 9.49453 -1.287775256 5.00E-05 0.0156975 XLOC_031828 Eln chr5:134702599-134747382 3.78599 8.52494 -1.17101908 5.00E-05 0.0156975 XLOC_037597 Calca chr7:114631479-114636357 4.2025 14.9264 -1.828546529 5.00E-05 0.0156975 XLOC_008866 Fkbp1b chr12:4833174-4841595 2.24346 9.86712 -2.136903593 5.00E-05 0.0156975 XLOC_041841 Scn10a chr9:119608455-119719032 0.176593 1.16162 -2.717638136 5.00E-05 0.0156975 XLOC_012587 Sncg chr14:34370273-34374789 8.16577 27.0804 -1.729588211 5.00E-05 0.0156975 XLOC_041842 Scn11a chr9:119753762-119825456 0.0733344 0.503005 -2.778010733 5.00E-05 0.0156975 XLOC_005730 Pirt chr11:66905720-66947086 0.424529 2.29628 -2.435363556 5.00E-05 0.0156975 XLOC_013755 Prph chr15:99055173-99058978 13.8156 35.4261 -1.358514432 5.00E-05 0.0156975 XLOC_005922 Tusc5 chr11:76679807-76698664 0.265702 1.96717 -2.888240646 5.00E-05 0.0156975 XLOC_035259 Slc17a6 chr7:51621829-51671125 0.323345 1.82905 -2.499948307 5.00E-05 0.0156975 XLOC_022041 Scn7a chr2:66673424-66784914 0.0834003 0.621018 -2.896510607 5.00E-05 0.0156975 XLOC_034297 Mgp chr6:136872435-136875805 2.09551 10.6148 -2.340703877 5.00E-05 0.0156975 XLOC_016887 Stub1 chr17:25773572-25988469 98.025 1244.89 -3.666724723 5.00E-05 0.0156975 XLOC_004173 Ppm1h chr10:122678693-122945795 7.33635 104.667 -3.834600375 5.00E-05 0.0156975 XLOC_044423 Lhfpl1 chrX:145290358-145349089 0 0.397838 inf 5.00E-05 0.0156975 XLOC_011907 Rps19-ps1 chr14:53213729-53214167 0 3.75537 inf 5.00E-05 0.0156975 XLOC_011872 Rps19-ps2 chr14:52902792-52903230 0 3.75537 inf 5.00E-05 0.0156975 XLOC_018884 Ifit1 chr19:34640870-34650009 1.2099 0.155232 2.962389917 0.0001 0.0272397 XLOC_006521 Rnf213 chr11:119390007-119487503 4.49868 1.82746 1.29966192 0.0001 0.0272397 XLOC_005839 Xaf1 chr11:72301593-72313733 2.8382 0.538327 2.398421564 0.0001 0.0272397 XLOC_040993 Acp5 chr9:22126730-22135746 9.3864 4.20928 1.156998461 0.0001 0.0272397 XLOC_040102 Thy1 chr9:44043194-44048579 3.69514 8.23127 -1.155486015 0.0001 0.0272397 XLOC_010851 Nrn1 chr13:36725624-36734477 0.355013 2.21826 -2.643484712 0.0001 0.0272397 XLOC_003053 Rgs4 chr1:169741476-169747642 3.26213 6.8039 -1.060547661 0.0001 0.0272397 XLOC_044134 Gm27733,Gm27927,XistchrX:103414466-103484957 0.0497755 10.6237 -7.737634789 0.00015 0.0330768 XLOC_040197 Isl2 chr9:55538671-55546665 0.127286 1.26589 -3.314006393 0.00015 0.0330768 XLOC_012314 Tpm3-rs7 chr14:113314607-113316713 25.1406 40.9918 -0.705316261 0.00015 0.0330768 XLOC_012188 Nefl chr14:68082589-68124899 12.298 26.9419 -1.131427884 0.00015 0.0330768 XLOC_018745 Mpeg1 chr19:12460639-12466000 30.08 19.0471 0.659233209 0.0002 0.0389372 XLOC_032588 Aqp1 chr6:55336431-55348555 0.735882 2.23271 -1.601249525 0.0002 0.0389372 XLOC_018829 Dock8 chr19:24999528-25202432 0.722477 0.283161 1.351329085 0.00025 0.0389372 XLOC_026592 N28178 chr4:42916659-42944752 0.739743 2.38391 -1.688233726 0.00025 0.0389372 XLOC_041725 Bsn chr9:108096021-108190384 2.28645 16.0498 -2.811374044 0.00025 0.0389372 XLOC_032213 Dync1i1 chr6:5725638-6028039 0.79517250 1.96754 -1.307054104 0.0003 0.0389372 XLOC_000421 Gm20257 chr1:58646687-58695989 8.20216 0.357011 4.521963455 0.00035 0.0389372 XLOC_029949 Pf4 chr5:90772434-90773381 9.98164 3.01072 1.72916833 0.00035 0.0389372 XLOC_005615 Igtp,Irgm2 chr11:58199555-58222779 2.77337 1.13239 1.292269187 0.00035 0.0389372 XLOC_008941 Rsad2 chr12:26442749-26456452 2.46628 0.623086 1.984833393 0.0004 0.0389372 XLOC_035193 Slc17a7 chr7:45163920-45176138 0.17048 0.752939 -2.14293049 0.0004 0.0389372 XLOC_039816 Gm10709 chr9:7751671-7752325 6.12769 1.96525 1.640630463 0.00045 0.0389372 XLOC_034116 C3ar1 chr6:122847137-122856161 7.16785 3.93681 0.864513361 0.00045 0.0389372 XLOC_018883 I830012O16Rik chr19:34607969-34613401 0.962846 0.124277 2.953745747 0.00055 0.0389372 XLOC_003130 Ifi204 chr1:173747292-173767047 1.57669 0.540157 1.545448332 0.0006 0.0389372 XLOC_015448 Parp14 chr16:35832873-35871544 1.38614 0.655534 1.080330462 0.00065 0.0389372 XLOC_024195 Ctss chr3:95526785-95556403 77.0373 50.4551 0.610557036 0.0007 0.0389372 XLOC_003129 AI607873 chr1:173723429-173741809 1.55442 0.645925 1.266937804 0.00085 0.0389372 XLOC_006006 Slfn5 chr11:82910549-82963791 1.05043 0.309098 1.7648438 0.0009 0.0389372 XLOC_040126 Scn4b chr9:45138436-45154152 0.0398219 0.329138 -3.047058637 0.0009 0.0389372 XLOC_011957 Trav4-4-dv10 chr14:53683646-53684177 0 0.447846 inf 0.00095 0.0389372 XLOC_012226 Lcp1 chr14:75131100-75230842 7.01281 4.08031 0.781313877 0.001 0.0389372 XLOC_004128 Ptprr chr10:116018212-116274932 0.800985 2.43439 -1.603713182 0.00105 0.0389372 XLOC_020153 Gm13430 chr2:36389131-36389419 0 0.636195 inf 0.00105 0.0389372 XLOC_010852 F13a1 chr13:36867177-37050244 2.46603 1.24057 0.991187208 0.00115 0.0389372 XLOC_040864 Lars2 chr9:123366939-123463153 147.553 102.48 0.525890873 0.00115 0.0389372 XLOC_021173 Hck chr2:153108467-153151441 2.13361 0.884802 1.26986994 0.00115 0.0389372 XLOC_019432 Pik3ap1 chr19:41271935-41385097 2.72654 1.28868 1.081177257 0.0012 0.0389372 XLOC_016463 Trem2 chr17:48346400-48354147 7.5047 3.59933 1.060066021 0.0012 0.0389372 XLOC_014211 Pdgfb chr15:79995899-80014808 0.960385 0.340438 1.49622079 0.00125 0.0389372 XLOC_042857 Tlr13 chrX:106143203-106160493 3.23148 1.69108 0.934250152 0.00125 0.0389372 XLOC_034144 Ptpn6 chr6:124720706-124738714 5.47428 2.48596 1.138866145 0.00135 0.0389372 XLOC_021610 Gm13321 chr2:12421626-12422026 0 0.303609 inf 0.0014 0.0389372 XLOC_038541 Hmox1 chr8:75093590-75100596 141.462 97.9775 0.529892178 0.00145 0.0402071 XLOC_009861 Gm2399 chr13:12702361-12702589 1.05605 0 inf 0.00145 0.0402071 XLOC_034925 Tyrobp chr7:30413787-30417577 51.0665 30.0388 0.765550002 0.00155 0.0429158 XLOC_019543 Hspa12a chr19:58795750-58860984 3.14219 5.1567 -0.714677697 0.0016 0.044234 XLOC_039127 Msr1 chr8:39581699-39642678 2.77372 1.13459 1.289651105 0.00165 0.0450117 XLOC_037357 P2ry6 chr7:100937637-100964391 2.95379 1.32776 1.153572866 0.0017 0.0463075 XLOC_001411 Arhgap30 chr1:171388953-171410298 1.33557 0.652351 1.033735266 0.00175 0.0475297 XLOC_034075 Ret chr6:118151747-118197744 0.502224 1.04396 -1.055663557 0.00175 0.0475297 Table 4.2 RNA -sequencing genes that were significantly altered in Gde3 KO
OPCs
51
Chapter 5 GDE3 regulates CNTF-CNTFRα signaling in proliferating OPCs
52
5.1 Introduction
Prior experiments have established GDE3’s role in regulating OPC proliferation, while RNAseq studies implicate misregulated cytokine signaling as a potential pathway governing Gde3 KO OPC hyperproliferation. We show here that CNTF-CNTFRα signaling is a potential mediator of GDE3-dependent inhibition of proliferation. This is because of the known roles for CNTF signaling in regulating OPC proliferation and that the ligand-binding component of the multiprotein CNTF receptor, CNTFRα, is a GPI- anchored protein abundantly expressed in OPCs. Further, GDE3 and CNTFRα are coexpressed in OPCs, and CNTFRα is a physiological substrate of GDE3. Here we test the involvement of the CNTF-CNTFRα pathway in GDE3-dependent control of OPC proliferation through multiple gain and loss of function approaches.
5.2 Results
5.2.1 CNTF inhibits OPC proliferation in a GDE3 dependent manner
Activation of CNTF signaling occurs through CNTFRα binding to CNTF; this results in a conformational shift that allows CNTF-CNTFRα to associate with the co- receptors gp130 and LIFRβ, generating a signaling receptor complex84. To test if GDE3 regulates OPC proliferation by modulating responsiveness to CNTF, we cultured isolated WT and Gde3 KO OPCs in the presence or absence of CNTF for 16 hours.
IdU was added to cultures 2 hours prior to fixation to identify cells in S-phase. In WT
OPCs, absence of CNTF results in an approximately a 32% increase in IdU incorporation, reflecting a marked increase in OPC proliferation (Figure 5.1E). Notably,
53 this increase is similar to the level of Gde3 KO OPC proliferation in the presence of
CNTF. In contrast to the WT condition, Gde3 KO cultures show no obvious changes when grown in the absence or presence of CNTF (Figure 5.1E). These observations suggest that CNTF works to reduce OPC proliferation via GDE3, and that the hyperproliferation of Gde3 KO OPCs may result from defective CNTF signaling.
5.2.2 Neutralization of CNTFRα-GP130 signaling increases OPC
proliferation
CNTF binds to CNTFRα which then recruits gp130 and LIFRβ to initiate downstream signal transduction cascades that include activation of MAPK, ERK, Akt and STAT signaling pathways84. To determine if the effects of CNTF on GDE3- mediated regulation of OPC proliferation occur via CNTFRα-gp130-LIFRβ signaling axis, we utilized two function blocking antibodies against gp130. The P8 antibody specifically blocks CNTFRα-gp130 signaling, and the R3 antibody blocks all gp130- mediated pathways (Figure 5.2A)91. Addition of P8 and R3 antibodies to WT cultures partially lifted the CNTF-dependent inhibition of proliferation by approximately 9% and
10%, respectively. In contrast, neither P8 nor R3 addition affected the proliferation of
Gde3 KO OPCs (Figure 5.2B). These experiments suggest that activation of CNTFRα- gp130-LIFRβ signaling constitutes one component of CNTF-GDE3 mediated pathways that inhibit OPC proliferation.
54
5.3 Summary
To determine the role of GDE3 in CNTF-CNTFRα signaling on OPC proliferation, we performed multiple manipulations on cultured WT and Gde3 KO OPCs including: 1) withdrawal of CNTFRα ligand, CNTF; and 2) addition of neutralizing antibodies against the CNTFRα-gp130-LIFRβ receptor complex. We show that cultured
WT OPCs display a pronounced decrease in proliferation upon addition of CNTF and that this effect is GDE3-dependent. Second, CNTF-mediated proliferation inhibits signaling, in part, through the CNTFRα-gp130-LIFRβ receptor complex. Taken together, these studies support the model that GDE3 regulates OPC proliferation through the regulation of CNTF-CNTFRα signaling.
55
Figure 5 .1 OPCs require GDE3 for CNTF -mediated suppression of proliferation (A -D’) Example images of WT and Gde3 KO OPCs cultured in the presence or absence of CNTF. IdU labeling highlights proliferative cells in S -phase. (E) Quantification of OPC proliferation in the presence (+) or absence ( -) of CNTF. Exclusion of CNTF from the culture media causes increased proliferation exclusively in WT OPCs (p* = 0.005, n =
9).
56
Figure 5.2 Neutralizing antibodies against gp130 increase WT OPC proliferation (A) Cartoon illustrating CNTF, CNTFRα, and co-receptors gp103 and LIFRβ. Function blocking antibodies against gp130 can inhibit its interaction with CNTF (P8) disrupting
CNTF signaling or LIFRβ (R3) which broadly disrupts cytokine transduction. (B)
Quantification of WT and Gde3 KO OPC proliferation in the presence of gp130 function blocking antibodies. In WT OPCs, P8 (p* = 0.001, n = 8) and R3 (p* = 0.0001, n = 8) antibodies significantly increased proliferation in the presence of CNTF.
57
Chapter 6 GDE3 mediates bimodal release of CNTFRα
58
6.1 Introduction
Previous studies show that GDE3 can release GPI-APs from the cell surface through cleavage of the GPI anchor. Our observation that GDE3’s control of OPC proliferation involves the CNTF-CNTFRα signaling pathway (Chapter 5) raises the possibility that this process involves GDE3 release of CNTFRα from the plasma membrane. This would be in line with previous studies showing that soluble forms of
CNTFRα can stimulate CNTF signaling. We utilize multiple cell culture and biochemical approaches to identify the nature of CNTFRα release by GDE3 and define its consequences for OPC proliferation.
6.2 Results
6.2.1 GDE3 releases CNTFRα via two separate mechanisms
Previous work from our laboratory and others reveal that six-transmembrane
GDEs can cleave the GPI-anchor that tethers proteins to the cell membrane, thus releasing them into the extracellular space51. Because GDE3 is capable of releasing
GPI-APs through this mechanism, we examined the consequences of GDE3 expression on CNTFRα release from the plasma membrane. We transfected plasmids expressing
GDE3 and CNTFRα into HEK293FT cells and analyzed the conditioned medium after
24 hours for the presence of CNTFRα by Western blot. The conditioned medium from
HEK293FT cells co-transfected with CNTFRα and GDE3 shows robust release of
CNTFRα, compared to when CNTFRα alone is transfected (Figure 6.1A). CNTFRα release by GDE3 is comparable to conditions including when cells transfected with
59
CNTFRα are treated with the bacterial protein PI-PLC, which cleaves within the GPI- anchor. However, unlike PI-PLC treated cells, medium from cells co-transfected with
GDE3 and CNTFRα contained an additional CNTFRα species that migrated at a slightly higher molecular weight (Figure 6.1A). This observation suggests that GDE3 is capable of releasing two different forms of CNTFRα into the medium. RNAseq analysis revealed that GDE3 ablation impacts the EV pathway. We examined the possibility that the larger, slower-migrating form of GDE3-released CNTFRα contains an intact GPI anchor tethering it to the EVs. We subjected the conditioned medium to phase separation using Triton X-114 detergent extraction. Cleaved, soluble, GPI-APs localize to the detergent poor fraction while uncleaved, membrane-bound, GPI-APs are enriched in the detergent rich fraction51. The higher molecular weight form of released CNTFRα partitioned to the detergent rich fraction, suggesting that it is released within a membrane bound compartment (Figure 6.1B). We further tested the hypothesis that
GDE3 releases CNTFRα in EVs by subjecting the conditioned media to established EV purification protocols involving high-speed centrifugation92. We validated successful EV purification through enrichment of EV-specific proteins CD9 and CD63 in the EV preparations. Strikingly, we found that CNTFRα was present in EV preparations when
GDE3 is present, but not when co-transfected with a vector control (Figure 6.3A). In summary, these observations suggest that GDE3 is capable of releasing CNTFRα through production of a soluble factor via GPI-anchor cleavage and as a membrane bound form in EVs.
60
6.2.2 Specific domains of GDE3 regulate CNTFRα release and other GPI-
APs
To determine whether GDE3 GPI-anchor cleavage activity and EV release capabilities are encoded by similar or distinct protein domains, we performed structure- function studies. GPI-anchor cleavage by the related family member GDE2 is dependent on key histidine residues located within the GDPD enzymatic domain52. To test if similar residues are required for GDE3 enzymatic activity, we mutated homologous histidine residues in GDE3 (H233A and H275A; GDE3.HHAA) (Figure
6.2A). GDE3.HHAA does not release the cleaved form of CNTFRα, however EV release is intact (Figures 6.3A). This observation confirms that the GPI-anchor cleavage function of GDE3 is located within the external GDPD domain, and it further suggests that EV release is governed by additional domains within GDE3. To identify regions required for GDE3-dependent release of EVs, we created a version of GDE3 without the intracellular N-terminus (GDE3.∆N) and a double mutant of GDE3 that is catalytically impaired and lacks the N-terminus domain (GDE3.HHAA∆N) (Figure 6.2A).
Surface biotinylation studies in HEK293FT cells confirm that all versions of GDE3 are expressed and traffic to the plasma membrane (Figure 6.2B). GDE3.∆N shows a complete loss of the ability to release CNTFRα in EVs and also reduced GPI-anchor cleavage; GDE3.HHAA∆N does not release CNTFRα in either form (Figure 6.3A).
These observations were confirmed using Triton X-114 fractionation of conditioned medium. Specifically, GDE3.HHAA retained the ability to release the higher molecular weight form of CNTFRα that fractionates to the hydrophobic detergent rich fraction
(Figure 6.3C). In contrast, GDE3.∆N released only the lower form of CNTFRα that
61 partitions to the hydrophilic detergent poor fraction (Figure 6.3B). GDE3.HHAA∆N does not release either form of CNTFRα. Taken together, these results indicate that GDE3’s ability to release EVs requires the N-terminal region while its enzymatic GPI-anchor cleavage activity requires the integrity of the external GDPD domain.
To examine if GDE3’s ability to release GPI-anchor proteins in EVs applies to other substrates we examined GDE3 dependent release of the GPI-linked proteins
RECK, CNTN1, GPC4, GPC6, CAD13 by sequential centrifugation of conditioned medium. These GPI-APs are released into the supernatant upon GDE3 co-expression
(Figure 6.4A) confirming involvement of GPI-anchor cleavage function; no release of substrates within the supernatant is observed in the presence of catalytically inactive
GDE3.HHAA. Strikingly, GDE3 is capable of releasing only a subset of GPI-APs by
EVs, specifically RECK, CNTN1 and GPC6. In line with this observation, GDE3.HHAA which normally retains EV release function, only releases RECK, CNTN1 and GPC6 while it does not release GPC4 and CAD13. Taken together, these observations suggest that GDE3 broadly releases GPI-AP substrates through GPI-anchor cleavage, but selectively releases GPI-AP substrates through EVs.
To test if GDE3 expression potentiates EV release, we performed nanoparticle tracking analysis of the HEK293FT cell conditioned media and assessed EV release irrespective of GPI-AP cargo. Strikingly, we find that expression of GDE3 induces a 15- fold increase in particles released compared to GFP control (Figure 6.5). Expression of
GDE3.HHAA produces a 6-fold increase in particles released. This observation suggests that GDE3’s enzymatic activity promotes EV release, but that it is not absolutely necessary. Notably, GDE3.∆N or GDE3.HHAA.∆N expression does not
62 stimulate EV release above control conditions. Therefore, the N-terminus of GDE3 is necessary for the generation of EVs. In summary, these findings suggest GDE3 is involved in inducing EV release from cells and this activity requires the N-terminus.
6.2.3 GDE3’s release of CNTFRα correlates with inhibition of OPC
proliferation
To examine the contribution of GDE3 differential release of CNTFRα to OPC proliferation, we compared the abilities of GDE3, GDE3.HHAA, GDE3.∆N and
GDE3.HHAA.∆N to inhibit OPC proliferation. We transfected constructs expressing these proteins in WT OPCs, cultured cells for 20 hours and pulse-labeled cells with IdU.
We then quantified the percentage of transfected OPCs that had incorporated IdU.
Expression of GDE3.HHAA suppressed the proliferation of OPCs to a similar extent as
WT GDE3, causing a 17% decrease in IdU uptake (Figure 6.6), while OPCs expressing
GDE3.∆N showed a 30% decrease in proliferation comparable to GDE3.HHAA. Of note, GDE3.HHAA∆N-expressing OPCs showed no change in proliferation compared to the vector control condition. These experiments suggest that GDE3 GPI-anchor cleavage and EV release mechanisms can independently act to inhibit OPC proliferation, and that abrogation of both activities is necessary to abolish GDE3 inhibitory function. The ability of GDE3 to inhibit OPC proliferation directly correlates with GDE3’s ability to release CNTFRα.
63
6.2.4 GDE3 is present on EVs and is enzymatically active
We find that GDE3 and GDE3.HHAA can release CNTFRα in EVs. Interestingly,
Western blot analysis of EV fractions, revealed that GDE3 and GDE3.HHAA are also present in EV fractions but not GDE3.∆N or GDE3.HHAA∆N (Figure 6.7A). This observation suggests the possibility that GDE3 and CNTFRα are co-released in EVs.
To test if GDE3 and CNTFRα are in the same EV, we transfected myc-tagged CNTFRα with GDE3 or GDE3.HHAA and performed coimmunoprecipitation on conditioned media with anti-myc magnetic beads. Analysis of the precipitates by Western blot detected both GDE3 and GDE3.HHAA (Figure 6.7B), suggesting that CNTFRα is co-released with GDE3 or GDE3.HHAA in EVs. This observation raises the possibility that GDE3 is co-released with CNTFRα in a signaling EV, where GDE3 could cleave and release
CNTFRα from the EV membrane. To test this possibility, we isolated conditioned medium from HEK293FT cells expressing CNTFRα and GDE3 or GDE3.HHAA. An aliquot from the media was immediately ultracentrifuged to purify EVs, while the remaining aliquots were incubated for 1.5 hours at 37°C prior to ultracentrifugation. The amount of CNTFRα in the EVs was then quantified by Western blot. The presence of
WT GDE3 resulted in a 71% decrease of EV-bound CNTFRα after 1.5 hours incubation, while there was no reduction observed in presence of catalytically impaired
GDE3.HHAA (Figure 6.7C and D). This suggests that GDE3 is capable of releasing
CNTFRα from EVs by GPI-anchor cleavage. We lengthened the incubation step to 24 hours and added a PI-PLC treatment to release the remaining EV bound CNTFRα.
After 24 hours, we observed a significant reduction of EV-bound CNTFRα in the presence of WT GDE3, but not GDE3.HHAA, consistent with continued release of
64
CNTFRα from EVs by GDE3 GPI-anchor cleavage (Figure 6.7D). Strikingly, addition of
PI-PLC caused a robust reduction of CNTFRα levels from EVs containing GDE3.HHAA but did not reduce levels of CNTFRα in EVs expressing WT GDE3 (Figure 6.7D), indicating that WT GDE3 had cleaved the majority of CNTFRα from the EVs by this timepoint. In summary, these observations suggest that GDE3 is co-released with
CNTFRα in EVs, where it can release CNTFRα from the EV surface membrane through
GPI-anchor cleavage.
6.2.5 Exogenous CNTFRα rescues hyperproliferation defects of Gde3 KO
OPCs
Thus far we have shown that GDE3 is capable of releasing CNTFRα through GPI anchor cleavage from the surface of cells and the surface of EVs. Previous studies have found that the addition of soluble versions of CNTFRα in the presence of CNTF is sufficient to activate the gp130-LIFRβ signaling axis93,94, suggesting that cleaved forms of CNTFRα can stimulate CNTF signaling. Our earlier studies showed that GDE3 inhibits OPC proliferation through stimulation of CNTF signaling. To determine if this occurs through release of soluble CNTFRα, we examined if soluble CNTFRα is sufficient to rescue the hyperproliferative phenotype of Gde3 KO OPCs. Commercially available recombinant CNTFRα (rCNTFRα) mimics soluble CNTFRα released by cleavage of the GPI-anchor. We incubated cultured Gde3 KO and WT OPCs with rCNTFRα in the presence of CNTF and PDGF for 18 hours and evaluated OPC proliferation by pulse labeling cultures with IdU. Addition of 1μg/ml rCNTFRα had no effect on the proliferation of WT OPCs cultures. In contrast, addition of the same
65 concentration of rCNTFRα to Gde3 KO cultures reduced OPC proliferation to WT levels
(Figure 6.7E). To test the sensitivity of Gde3 KO OPCs to soluble rCNTFRα, we repeated the experiment but decreased the concentration 100-fold to 0.01μg/ml. This lower concentration of rCNTFRα was sufficient to reduce proliferation of Gde3 KO
OPCs. Addition of rCNTFRα to WT cultures at either 1μg/ml or 0.01μg/ml had no effect on OPC proliferation. These observations suggest that OPC proliferation of Gde3 KO
OPCs is highly sensitive to soluble forms of CNTFRα, and supports the model that
GDE3 regulates OPC proliferation by stimulating CNTF signaling through release of
CNTFRα.
6.3 Summary
We show here that GDE3 is capable of releasing CNTFRα by two distinct mechanisms, specifically through cleavage at the GPI-anchor and as an EV bound form. Structure-function studies identify distinct domains within GDE3 protein that are responsible for the dual release modalities, and demonstrate that both functions are required for regulating OPC proliferation. Further analysis of EVs reveals that GDE3 is co-released with CNTFRα in EVs, where it remains enzymatically active to cleave
CNTFRα off the EV membrane. Lastly, we show that addition of soluble rCNTFRα is sufficient to rescue Gde3 KO OPC hyperproliferation. These observations identify new properties of GDE3 governing the release of GPI-APs in EVs and delineate a mechanism by which GDE3 regulates OPC proliferation via stimulation of CNTF signaling through release of CNTFRα (Figure 6.8).
66
Figure 6.1 GDE3 releases two species of CNTFRα. (A) HEK293FT cell assay shows that GDE3 releases two forms of CNTRFα, a larger
(100 kDa) (*) and a smaller (90 kDa) (**) form. Forced GPI-anchor cleavage by PI-PLC only releases the smaller form of CNTFRα. The 70 kDa version of CNTFRα seen in lysate is not released by GDE3. Separate lanes from the same blot are adjoined for clarity. (B) Fractionation of medium with Triton X-114 into detergent rich (membrane bound) and detergent poor (non-membrane bound) separates the two forms of
CNTFRα released by GDE3.
67
Figure 6. 2 Mutant versions of GDE3 are effectively processed and trafficked to the cell surface. (A) Topology diagrams of WT and mutant GDE3 constructs used for transfection. (B)
Western blot following surface biotinylation in HEK293FT cells expressing empty vector, wildtype GDE3, or mutant GDE3. All versions of GDE3 are effectively glycosylated (doublet indicates different degrees of glycosylation) and are effectively trafficked to the plasma membrane (present in the biotinyla ted fraction).
68
Figure 6. 3 GDE3 releases soluble and membrane bound CNTFRα. (A) Ultracentrifugation of HEK293FT cell assay medium separates soluble proteins
(supernatant) and EVs (pellet). The EV pellet is positive for EV mark ers CD9 and CD63 and contains CNTFRα released by GDE3 and GDE3.HHAA. CNTFRα released by
GDE3.ΔN is restricted to the supernatant. GDE3.HHAAΔN is undistinguishable from empty vector control in both fractions. CD9 and CD63 are run under non -reducing conditio ns. (B and C) HEK293FT cell assay and Triton X -114 fraction with GDE3 mutants. The larger form (*) released by GDE3.HHAA partitions to the detergent rich fraction and the smaller form (**) released by GDE3.ΔN separates into the detergent poor fraction. Rel ease of membrane -bound CNTFRα by GDE3 is more robust without the 70 minute interval needed for ultracentrifugation (compare panel A)
69
Figure 6.4 GDE3 release of GPI-APs. (A) HEK293FT cell assays transfecting empty vector or constructs expressing versions of GDE3 along with the indicated HA-tagged GPI-anchored protein. Samples of medium were ultracentrifuged to separate cleaved GPI-AP (supernatant) and EV-bound GPI- anchored protein in pellets. Selective GPI-anchored substrates are predominantly cleaved (GPC4 and CAD13) while others are cleaved and released into EVs (RECK and CNTN1).
70
Figure 6.5 Evaluation of GDE3 mediated release of extracellular particles. (A) Nanoparticle tracking of EVs released during the HEK293FT cell assay.
Expression of GDE3 yields a 15.16 fold increase in EV release (p* = 0.001, n = 6) compared to GFP control. GDE3.HHAA produces a smaller 6.42 fold increase (p* =
0.001, n = 6) in EVs, and GDE3.ΔN and GDE3.HHAAΔN do not increase EV release.
71
Figure 6.6 Proliferation assay of transfected OPCs with GDE3 constructs (A) Graph representing the degree of IdU incorporation upon transfection of OPCs with a given GDE3 construct normalized to GFP -only control. Significant inhibition of prol iferation occurs upon transfection of GDE3 (p* < 0.001, n = 12), GDE3.HHAA (p* =
0.036, n = 12), and GDE3.ΔN (p* < 0.001, n = 12); however, proliferation is unaffected following transfection of GDE3.HHAAΔN.
72
73
Figure 6.7 Non-membrane bound CNTFRα is sufficient to inhibit OPC proliferation. (A) GDE3 and GDE3.HHAA protein are released from HEK293FT cells in EVs. (B) Co-
IP analysis demonstrates that GDE3 and released CNTFRα are released in the same
EVs. (C and D) Western blot and quantification of CNTFRα release from EVs.
Incubation of EVs containing GDE3 and CNTFRα for 1.5 hours shows depletion
(release) of CNTFRα from the EV fraction (p* = 0.01, n = 4). Additional release is detectable from 1.5 hours to 24 hours (p* = 0.027, n = 4). No change is detectable with
GDE3.HHAA co-expression. After 24 hour incubation, addition of PI-PLC significantly reduces the level of CNTFRα in GDE3.HHAA EVs (p* < 0.001, n = 5) but has no effect on WT GDE3 EVs. (E) Quantification of proliferation following addition of recombinant
CNTFRα in WT and Gde3 KO OPCs. Recombinant CNTFRα has no effect on WT OPC proliferation but significantly reduces Gde3 KO proliferation at concentrations of
0.01µg/ml (p* = 0.025) and 1µg/ml (p* = 0.023).
74
Figure 6.8 Working model for GDE3 dependent regulation of OPC proliferation. GDE3 induces the release of CNTFRα via GPI-anchor cleavage (dashed red lines) or
EV release (dashed blue line). Cleavage is dependent on histidine residues in the
GDPD domain, and EV release is dependent on the N-terminus. Cleaved CNTFRα binds CNTF and signals through gp130 and LIFRβ to inhibit OPC proliferation.
75
Chapter 7 Summary and Future Directions
76
7.1 Summary
GDE3 is expressed within the ventricular zone of the developing spinal cord encompassing the pMN domain from which motor neurons and subsequently oligodendrocytes originate. Previous studies of GDE3 in osteoblast cells suggest that
GDE3 negatively regulates cellular proliferation56. These observations raise the possibility that GDE3 negatively regulates the proliferation of cells within the ventricular zone, including OPCs. To test this hypothesis, we performed genetic loss of function studies and showed that loss of GDE3 affects the cell-cycle kinetics of glial progenitors, by increasing proliferation and shortening cell cycle length. To assess the consequences of Gde3 loss on OPCs specifically, we utilized a mosaic mouse model that allows direct comparison between WT and Gde3 KO OPCs, and we observed that there was an increased number of Gde3 KO OPCs that subsequently differentiated into mature oligodendrocytes. Analysis of purified cortical and spinal cord showed that
OPCs displayed heightened proliferation and accelerated differentiation in absence of
GDE3, akin to our in vivo models. Expression of GDE3 in WT OPCs is sufficient to inhibit proliferation, and GDE3 acts in a cell autonomous manner in this capacity.
Taken together, these observations suggest that GDE3 is required to regulate the pace of OPC differentiation through cell-autonomous negative regulation of OPC proliferation and differentiation.
To elucidate the molecular mechanism(s) underlying GDE3 function, we performed RNA-sequencing on isolated WT and Gde3 KO OPCs to evaluate global transcriptional changes elicited by loss of GDE3. Functional enrichment analysis of differentially expressed genes identified several misregulated pathways, including
77 extracellular vesicles and cytokine signaling pathways. We hypothesize that the cytokine signaling CNTF pathway and its receptor GPI-AP CNTFRα are putative mediators of GDE3-dependent inhibition of OPC proliferation. This hypothesis is based on previous observations that this pathway is involved in OPC proliferation, and that
CNTFRα38,43,47,95, the ligand binding component of the CNTF receptor, is a GPI- anchored protein84. We assessed the involvement of the CNTF-CNTFRα in GDE3- dependent control of OPC proliferation through gain and loss of function approaches.
We found that WT OPCs display decreased proliferation upon addition of CNTF, and that this phenotype is dependent on signaling through the CNTFRα-gp130-LIFRβ receptor complex and requires GDE3.
Previous studies reveal that CNTF signaling can be mediated by released
CNTFRα94. We found that Gde3 KO animals show reduced release of CNTFRα, suggesting that GDE3 regulates CNTF-mediated inhibition of OPC proliferation through regulated release of CNTFRα. We show that GDE3 releases CNTFRα as a soluble factor via GPI-anchor cleavage and also in a membrane-bound form via EVs.
Structure-function studies reveal that two catalytic histidine residues within the extracellular enzymatic domain are required to generate soluble CNTFRα, while the intracellular N-terminus is necessary for CNTFRα EV release. We find GPI-anchor cleavage and EV release mechanisms act independently to inhibit OPC proliferation, and that abrogation of both activities is obligatory for abolition of GDE3 inhibitory function. Our biochemical analyses detect GDE3 in EVs where it remains enzymatically active and can release CNTFRα from the membrane by means of GPI-anchor cleavage, in line with the formation of a signaling exosome. These observations suggest that
78 generation of soluble CNTFRα is an important component of GDE3 stimulation of CNTF signaling to mediate inhibition of OPC proliferation. Consistent with this idea we found that addition of soluble recombinant CNTFRα to OPC cultures was sufficient to rescue
Gde3 KO OPC hyperproliferation.
Based on these results we propose a model whereby GDE3 regulates OPC proliferation through CNTF-dependent signaling. GDE3 releases CNTFRα from the membrane surface via GPI-anchor cleavage from the cell surface, and also by a novel mechanism that involves recruitment of CNTFRα into EVs prior to its release from EVs via cleavage of the GPI-anchor. Thus, GDE3 employs bi-modal release mechanisms to negatively regulate OPC proliferation through stimulation of CNTF signaling.
7.2 Discussion and Future Directions
7.2.1 GDE3 function in OPC proliferation and differentiation
This study reveals several functions for GDE3 during central nervous system development. In the developing spinal cord, onset of GDE3 expression coincides with the ‘neuron-glial’ switch, and GDE3 expression is restricted to glial cell lineages, suggesting functions in glial development. While GDE3’s expression encompasses the majority of ventral progenitors, we specifically focus here on OPCs derived from the pMN domain. OPCs lacking GDE3 exhibit increased proliferation and higher propensity for differentiation. Our studies suggest GDE3 constitutes a regulatory module to slow down OPC development, and we speculate that this mechanism operates to prevent depletion of the oligodendrocyte progenitor pool and also to coordinate with neuronal development for synchronization of axon myelination. In the adult, OPCs form a
79 quiescent population that balances growth with self-repulsion to achieve spacing homeostasis96. This resident OPC population reacts through proliferation and differentiation in response to injury or insult9. It would be of great interest to determine if
GDE3 inhibitory function extends to maintain this homeostasis of resident OPCs in the adult nervous system. GDE3’s ability to regulate proliferation is also observed in other systems. Expression of GDE3 is sufficient to inhibit osteoblast proliferation in cultured cell lines, and GDE3 overexpression slows tumor growth in xenograft mouse models56,57. Our study, in conjunction with above-mentioned observations, corroborates the idea that GDE3 is an important regulator of cellular proliferation.
Our studies also suggest that GDE3 regulates the pace of OPC proliferation and their maturation into myelinating oligodendrocytes. The ability of GDE3 to both inhibit
OPC proliferation and differentiation is unique. While we find that GDE3 modulates the
CNTF signaling pathway to regulate OPC proliferation, we have not explored the mechanism behind GDE3’s inhibition of OPC differentiation. It is possible that CNTFRα also acts to promote differentiation, however GDE3 may exert its influence on other
GPI-APs such as Contactin-1, which has been reported to regulate maturation of OPCs and myelination of differentiated oligodendrocytes26. Further studies are required to establish GDE3’s mechanism for inhibition of differentiation of OPCs.
7.2.2 GDE3 regulation of CNTF signaling
We have identified the CNTF signaling pathway as one mechanism by which
GDE3 regulates OPC proliferation. We find that the addition of CNTF to WT OPCs normally inhibits their proliferation; however, this inhibition is abolished in the absence
80 of GDE3. Furthermore, neutralizing antibodies that block components of the CNTF- mediated signaling via the CNTFRα-gp130-LIFRβ signaling complex relieve CNTF- dependent inhibition but do not affect Gde3 KO OPC proliferation. We propose that
CNTF signaling mediated through the CNTFRα-gp130-LIFRβ complex inhibits OPC proliferation, and that this pathway is regulated via GDE3. Our observations indicate that GDE3 mediates CNTF-dependent OPC proliferation by release of CNTFRα. This raises the question; what is the benefit of releasing CNTFRα from the cell surface of
OPCs considering that CNTFRα is already present on OPC plasma membrane?
Previous studies have shown that soluble versions of CNTFRα can stimulate CNTF signaling. Thus, it is possible that released CNTFRα is more effective in transducing
CNTF-mediated signaling than membrane-tethered isoforms. In addition, it is conceivable that releasing CNTFRα may be a mechanism by which accessibility to other components of the CNTF receptor can be readily achieved. In this scenario,
GDE3 release of soluble CNTFRα could facilitate its association with gp130-LIFRβ complexes that are distributed throughout the cell surface. Indeed, studies have revealed that presence of the CNTF receptor in different regions of the plasma membrane can stimulate different downstream signaling pathways97. We speculate that increased accessibility of released CNTFRα to remote locations of gp130-LIFRβ on the
OPC plasma membrane stimulates relevant downstream pathways required for inhibition of OPC proliferation.
CNTF is a pleotropic cytokine reported to affect a variety of cell types that in turn regulate a wide range of effects including cell survival, proliferation and differentiation38,43,46,89,98. Our studies identify functions for CNTF in inhibiting OPC
81 proliferation, and this discovery is in line with roles for CNTF in inhibiting cellular proliferation in other systems such as chick sympathetic neural progenitors, osteoblasts and neural crest cells99–101. However, other studies have shown that CNTF can also stimulate cellular proliferation, for example in myoblasts and Muller glia derived progenitors44,95. How could this occur? Studies in cell lines reveal that CNTF effects on proliferation are biphasic, where stimulation or inhibition of cellular proliferation depends on the concentration of the ligand102. Therefore, CNTF effects on OPC proliferation likely reflect a highly tuned response that integrates multiple components, including local ligand availability.
Our analyses of GDE3 expression in OPC cultures and HprtGFP animals suggests that GDE3 functions cell-autonomously to regulate OPC proliferation. In cultures transfected with GDE3, only transfected OPCs show inhibition of proliferation, while non-transfected cells in the same culture do not. In addition, Gde3 KO OPCs show increased proliferation in HprtGFP animals when compared to WT OPCs in the same animal. We note that GDE3 and CNTFRα are both expressed in astrocytes, raising the possibility that CNTFRα released by GDE3 in astrocytes could also contribute to OPC proliferation. Although our current studies do not rule out this possibility, this contribution is likely to be minor. In the HprtGFP model, Gde3 KO and
WT OPCs reside within the same environment that also contains WT and Gde3 KO astrocytes. If released CNTFRα from astrocytes plays a major part in OPC inhibition, then no differences would be detected between WT and Gde3 KO OPCs proliferation.
Taken together, our observations argue that GDE3 function in OPCs proliferation is cell autonomous.
82
7.2.3 GDE3 and EV release
Our laboratory and others have previously defined the GPI-anchor cleaving ability of GDE357, and GDE3 is also reported to directly metabolize glycerophosphoinositol by an analogous PLC-cleavage mechanism56. We identify here a novel function of GDE3 as a potent releaser of EVs. GDE3-induced EV release is several orders of magnitude higher than published pharamacological and overexpression-based manipulations that trigger EV production103–105. The GDE3 N- terminus domain is essential for EV release and is distinct from the extracellular enzymatic region required for GDE3 GPI-anchor cleavage and glycerophosphoinositol metabolism. Nanoparticle tracking analysis suggests partial involvement of GDE3 enzymatic activity since GDE3.HHAA mutants show reduced EV release capability.
Together, these observations suggest that GDE3 regulates EV release through two different mechanisms. Whether GDE3 utilizes known ESCRT and ceramide-based pathways or a novel mechanism is unknown; however, our observation that GDE3 promotes robust release of EVs without commensurate increases in the known EV markers CD9 and CD63 is consistent with the latter. Identification of GDE3 binding partners at the plasma membrane would provide insight into the mechanisms of GDE3- dependent release of EVs. We note here that GDE3ΔN shows reduced ability to cleave
GPI-APs. GDE3’s N-terminus contains a potential N-myristoylation site and possesses
6 cysteines predicted to become S-palmitoylated (Table 7.1)106. Both modifications increase hydrophobicity of the protein, thus increasing preference for cholesterol enriched membrane microdomains107, where GPI-APs are often preferentially
83 localized108. S-palmitoylation is a reversible process and so could provide a regulatory function for GDE3-dependent signaling mechanisms. We hypothesize that the reduced
GPI-anchor cleavage activity of GDE3ΔN is because GDE3ΔN fails to colocalize with
GPI-APs within the appropriate plasma membrane microdomain. Further experiments on GDE3’s N-terminus involving mutagenesis of potential lipid modified amino acids would test this hypothesis.
Our studies in heterologous cells reveal that GDE3 does not release all GPI- anchored proteins through bi-modal mechanisms. Certain substrates such as GPC4 and CAD13 are primarily released by GDE3 GPI-anchor cleavage and not by EVs
(Figure 6.5). GPI-anchors are heterogenous structures modified by additions of saccharides and fatty acids109. These have been shown to regulate GPI-AP localization and binding properties110. The basis for GDE3’s selective recruitment of GPI-AP into
EVs is currently unknown but may be related to protein structure, GPI-anchor structure, or protein localization.
Our studies suggest that GDE3 recruits CNTFRα into EVs where it can release
CNTFRα by GPI-anchor cleavage. The formation of a signaling EV containing GDE3 and CNTFRα has additional benefits to cleavage and release from the cell surface. EV cargoes are highly stable. Accordingly, the signaling EV could constitute a stable reservoir of GDE3 and CNTFRα that could be rapidly activated to provide a source of released CNTFRα to bind CNTF and initiate co-receptor activation on the plasma membrane. It remains unclear whether GDE3-dependent cleavage of CNTFRα is regulated, and it will be of interest to explore this possibility further. We note that the
GPI-anchor cleavage activity of the closely related family member GDE2, is tightly
84 regulated by thiol-redox mechanisms, and it is possible that similar thiol-redox control mechanisms apply to GDE352. We speculate that the GDE3-dependent formation of signaling EVs could be applied to other GPI-AP substrates and operate in other tissues where GDE3 is expressed such as spleen, lungs and bone. Our study provides insight into the mechanism of GDE3 function and raises the possibility that bimodal release of
GPI-anchored proteins is central to six-transmembrane GDE3 protein function in multiple cellular contexts.
85
Table 7.1 GPS-Lipid analysis of GDE3
Analysis of GDE3 protein by GPS-Lipid106 predicts several post-translational lipid modifications on amino acids highlighted in red. N-terminus of GDE3 (Positions 1-40) is potentially modified by N-myristoylation and 6 palmitoylations.
86
Materials and Methods
87
HEK293FT Maintenance and Transfection
HEK293FT cells were maintained in DMEM (Invitrogen) plus 10% fetal bovine serum (Sigma) and 1% penicillin-streptomycin (Life Technologies) in a 37°C incubator with 5% CO2. For transfections, 12 well plates were coated with 25µg/ml polyethyleneimine (PEI) in 150mM NaCl for 1 hour at 37°C and then washed 3 times with PBS. HEK293FT cells were plated on PEI coated plates at a density of 150,000 cells per well (37,500 cells per cm2) in 1mL of DMEM+FBS. One day after plating, cells were transfected with FuGENE HD (Promega) following manufacturer's instructions with the indicated plasmids. One day following transfection, media was replaced with fresh
DMEM and conditioned for 16 hours unless otherwise noted. Where indicated, PI-PLC
(Invitrogen) was added to media 1 hour prior to analysis.
Media and Tissue Fractionation
For media fractionation using ultracentrifugation, conditioned medium plus protease inhibitors was spun at 3,000xg for 10 minutes at 4°C to remove cellular debris.
Extracellular vesicles were pelleted at 100,000xg for 70 minutes at 4°C in a fixed angle
TLA 100.4 rotor. The pellet was resuspended in either SDS loading buffer for SDS
PAGE and immunoblotting or PBS for particle tracking. Particle detection and quantification were performed on a ZetaView Nanoparticle Analyzer (ParticleMetrix) following the manufacturer’s protocol. Triton X-114 phase partitioning was performed as previously described51. Briefly, 2% Triton X-114 (Sigma) was purified by mixing 1:1 with
Tris Buffer (100mM Tris-HCl pH 7.4, 150mM NaCl). This solution was phase separated by raising the temperature to 30°C for 5 minutes. The buffer was aspirated, and this
88 process was repeated twice. For media fractionation purified 2% Triton X-114 was added 1:1 with the medium and incubated at 4°C for 10 minutes with occasional vortexing. For tissue fractionation, cortex of P2-P4 mice was sonicated in purified 2%
Triton X-114 and spun at 4C 21,000xg to remove debris. Aliquot of input was saved and the rest was used for partitioning. The detergent-rich pellet and detergent-free supernatant were separated by incubating at 30°C for 5 minutes followed by centrifugation at 3,000xg for 3 minutes.
Surface Biotinylation
Primary OPCs or transfected HEK293FT cells were chilled on ice and all solutions were cooled on ice. Cells were washed with PBS three times. Freshly made sulfo-NHS-SS-biotin solution was added for 30 minutes with gentle rocking every 5 minutes. Cells were then washed once with PBS and then quenched with 100mM glycine solution. Cells were washed with PBS and lysed with RIPA (PBS, 1% Triton X-
100, 0.5% sodium deoxycholate, 0.1% SDS) buffer containing protease inhibitors and rocked for 20 minutes at 4C. The lysate was sonicated and spun at 21,000xg for 20 minutes at 4°C to remove any debris. A fraction was saved as input and the remainder of the supernatant was mixed with avidin-agarose beards and rotated at 4°C overnight.
The beads were washed with RIPA buffer, and proteins were eluted by addition of SDS loading buffer plus 10% β-mercaptoethanol and heated at 60°C for 15 minutes.
89
EV Cleavage Assay and Co-Immunoprecipitation
HEK293FT cells were plated on 100mm PEI coated plates at a density of 1.5x106 cells per plate. Cells were transfected with FuGENE HD following manufacturer’s instructions. Following transfection, medium was switched to DMEM and conditioned for
3 hours. Medium was transferred to a tube containing protease inhibitors and spun at
3,000xg for 10 minutes at 4°C to remove cellular debris. The medium was then aliquoted into individual tubes and incubated in a 37°C water bath for the specified amounts of time, followed by centrifugation. Where indicated, PI-PLC (Invitrogen) was added for 1 hour prior to ultracentrifugation. Pellets were resuspended in SDS- containing loading buffer. For co-immunoprecipitation experiments, HEK293FT cell conditioned medium was incubated with anti-myc magnetic beads (Thermo) to isolate myc-CNTFRα. Beads were washed with TBS + 0.15% Tween-20 prior to elution with
SDS-containing loading buffer.
OPC Isolation and Proliferation Assays
OPCs were isolated from P2-4 cortices or spinal cords as indicated using Miltenyi
Biotec magnetic beads for A2B5+ cells according to manufacturer’s recommendations.
OPCs were plated on Poly-L-Lysine (0.1mg/mL) and Laminin (10µg/mL) coated plates at a density of 150,000 cells per cm2 in proliferative media containing 1mM Sodium
Pyruvate, 5µM Forskolin, 10ng/mL PDGF, 10ng/mL CNTF, B27, N2, and antibiotic unless otherwise specified. OPC medium was made with freshly reconstituted growth factors, 0.22um filtered and stored at 4°C for no longer than 3 days. For transfections,
Lipofectamine 2000 (Invitrogen) was used according to manufacturer’s
90 recommendations. Briefly, OPCs were cultured in proliferative medium for at least 16 hours prior to transfection. Transfection was performed for 1 hour in MEM (Gibco). Cells were washed once with OPC proliferation medium without antibiotic and incubated for at least 16 hours prior to the proliferation assay.
For all OPC proliferation assays, IdU was added (20µg/mL) for 2 hours prior to fixation with 4% paraformaldehyde in 0.1M Phosphate Buffer (PB) for 15 minutes at room temperature. To visualize IdU, antigen retrieval was performed by treating the cells with 2N HCl for 7 minutes at 37°C. Cells were washed with PBS and neutralized in
0.1M Borate buffer. Where indicated, function-blocking antibodies P8 or R3 (Abcam, 10
µg/mL) were included throughout the course of the OPC culture. For differentiation assay, OPCs were maintained in presence of 10ng/mL PDGF for 1 day following isolation. Afterwards, OPCs were induced to differentiate by change of OPC media without PDGF and incubated for three days prior to fixation and immunofluorescence analysis.
Animal Husbandry and Tissue Preparation
Mice were maintained and used in accordance with approved Johns Hopkins University
IACUC protocols. Mice were anesthetized and transcardially perfused with 0.1M PB followed by 4% paraformaldehyde (PFA) in 0.1M PB. Tissue samples were post-fixed in
4% PFA for 1 hour at 4°C, washed in PBS, and incubated in 30% sucrose for 24 hours.
Samples were embedded in O.C.T Compound (Tissue-Tek 62550–12) and flash frozen in a dry ice ethanol bath. Cryomolds were stored at −80°C and sectioned on an UltraPro
5000 Cryostat (Vibratome).
91
Immunocytochemistry
Tissue sections were air dried for at least 20 minutes then washed twice in PBS.
Both tissue sections and cultured cells were blocked in PBS with 0.3% Triton X-100 and
10% heat inactivated normal goat serum (HINGS) for 30 minutes. Primary antibody incubation was performed in PBS with 0.3% Triton X-100 and 10% HINGS at 4°C overnight. Primary antibodies used were as follows: rabbit anti-Ki67 (1:1,000, Abcam, cat. no. ab15580), guinea pig anti-Nkx6.1 (1:4,000, from T.M. Jessell), guinea pig anti-
Olig2 (1:20,000, from B. Novitch), guinea pig anti-NG2 (1:500, from D. Bergles), rat anti-
PDGFRα (1:200 BD Sciences 558774), Rabbit anti-Cleaved Caspase 3 (1:1000, Cell
Signaling Technologies 9661S). Samples were washed in PBS and incubated with secondary antibodies (Jackson ImmunoResearch) for 1 hour at room temperature.
Samples were stained with DAPI (Invitrogen R37606). Slides were coverslipped with
Vectashield (Vector Laboratories) mounting medium and imaged using a Zeiss LSM
700 or Keyence BZ-X710 microscope. Brightness and contrast are adjusted evenly between experimental groups.
Immunoblotting
Samples were resolved using SDS-PAGE, transferred to PVDF membrane, blocked with 5% milk and probed with indicated antibodies overnight at 4°C: CD9 (1:500
Biolegend 312102), CD63 (1:500 Santa Cruz sc-5275), FLAG (1:10,000 Sigma F7425),
Myc (1:10,000 Cell Signaling Technologies 2276), HA (1:1,000 Cell Signaling
Technologies 2367S), RECK (1:1000 Cell Signaling Technologies 3433), HRP-
92 conjugated secondary antibodies (Jackson ImmunoResearch). Samples probed for CD9 and CD63 were lysed in non-reducing conditions. Membranes were developed with
Western Blot Detection Kit (Kindle Biosciences) and imaged on autoradiography paper or digital imaging (KwikQuant, Kindle Biosciences). Bands were quantified using
ImageJ.
RNA Sequencing
Spinal cords from P0.5 pups were dissected and the OPCs were isolated and cultured as described above. One day following isolation RNA was extracted from cells using
RNeasy Plus Micro Kit (Qiagen). cDNA libraries were prepared using the Illumina
TruSeq Stranded mRNA Library Prep Kit (Illumina, RS-122-2101). Paired-end reads, 50 bp in length, were generated on an Illumina HiSeq 2500 system. To analyze the RNA- seq data, reads were quality checked and trimmed using the programs fastqc and fqtrim. Reads were then mapped to the mouse genome mm10 using the spliced alignment program Tophat2 v2.1.1, and assembled into transcripts using Cufflinks v2.2.1. Transcript assemblies across all samples were merged with Cuffcompare v.2.2.1, using GENCODE v.M5 as reference, to create a set of gene and transcript annotations that was later used in the differential analyses. Lastly, Cuffdiff v2.2.1 was run on each pairwise comparison to determine statistically significant differentially expressed genes (significance cutoffs: p-value <= 0.05, q-value <= 0.05)
93
Quantification and Statistical Analysis
Image quantification was performed with ImageJ. All quantification was performed using raw data while blind to the experimental condition. For in-vivo tissue sections, a minimum of 10 sections were quantified per embryo. Embryos were generated from a minimum of 2 litters for each experiment. Regions of interest were determined with the relevant counterstain, either DAPI or indicated VZ marker. For in- vitro cell counts, a minimum of 3 wells were quantified per animal, and at least 3 animals were used for each experiment. The total number of cells quantified for the in- vitro experiments is documented in Table S1. S-phase measurements and calculation of cell cycle length was performed as previously described85. Briefly, BrdU was injected
(70mg/kg by weight) intraperitoneally into pregnant dams 2 hours prior to sacrifice. IdU was injected using an equimolar dose 30 minutes before sacrifice. Counts were performed within the ventral region of the spinal cord identified by Nkx6.1 positive cells.
Graphs represent the mean ± SEM. The reported n number refers to individual animals, processed uniformly across experimental conditions. Statistical significance was determined using a two-tailed, unpaired Student’s t test; except in Figure 6D where the analysis was paired. Significance level is a value of p < 0.05.
94
References
95
1. Jessell, T. M. Neuronal Specification in the Spinal Cord : Inductive Signals and
Transcriptional Codes. 1, 20–29 (2000).
2. Kintner, C. Neurogenesis in Embryos and in Adult Neural Stem Cells. J. Neurosci.
22, 639–643 (2002).
3. Rowitch, D. H. & Kriegstein, A. R. Developmental genetics of vertebrate glial-cell
specification. Nature 468, 214–222 (2010).
4. Corda, D., Mosca, M. G., Ohshima, N., Grauso, L. & Yanaka, N. The emerging
physiological roles of the glycerophosphodiesterase family. 281, 998–1016
(2014).
5. Seidl, A. H. Reviewregulation of Conduction Time Along Axons. Neuroscience
276, 126–134 (2014).
6. Lin, S. C. & Bergles, D. E. Synaptic signaling between neurons and glia. Glia 47,
290–298 (2004).
7. Stolt, C. C. et al. The Sox9 transcription factor determines glial fate choice in the
developing spinal cord. Genes Dev. 17, 1677–1689 (2003).
8. Deneen, B. et al. The Transcription Factor NFIA Controls the Onset of
Gliogenesis in the Developing Spinal Cord. Neuron 52, 953–968 (2006).
9. Bergles, D. E. & Richardson, W. D. Oligodendrocyte development and plasticity.
Cold Spring Harb. Perspect. Biol. 8, (2016).
10. Calver, A. R. et al. Oligodendrocyte population dynamics and the role of PDGF in
vivo. Neuron 20, 869–882 (1998).
11. Fruttiger, M. et al. Defective oligodendrocyte development and severe
hypomyelination in PDGF-A knockout mice. Development 126, 457–467 (1999).
96
12. McKinnon, R. D., Matsui, T., Dubois-Dalcq, M. & Aaronsont, S. A. FGF modulates
the PDGF-driven pathway of oligodendrocyte development. Neuron 5, 603–614
(1990).
13. Frederick, T. J. & Wood, T. L. IGF-I and FGF-2 coordinately enhance cyclin D1
and cyclin E-cdk2 association and activity to promote G1progression in
oligodendrocyte progenitor cells. Mol. Cell. Neurosci. 25, 480–492 (2004).
14. Deverman, B. E. & Patterson, P. H. Exogenous Leukemia Inhibitory Factor
Stimulates Oligodendrocyte Progenitor Cell Proliferation and Enhances
Hippocampal Remyelination. J. Neurosci. 32, 2100–2109 (2012).
15. Linker, R. A. et al. CNTF is a major protective factor in demyelinating CNS
disease: A neurotrophic cytokine as modulator in neuroinflammation. Nat. Med. 8,
620–624 (2002).
16. Miyamoto, Y., Yamauchi, J. & Tanoue, A. Cdk5 Phosphorylation of WAVE2
Regulates Oligodendrocyte Precursor Cell Migration through Nonreceptor
Tyrosine Kinase Fyn. J. Neurosci. 28, 8326–8337 (2008).
17. Bribián, A., Barallobre, M. J., Soussi-Yanicostas, N. & de Castro, F. Anosmin-1
modulates the FGF-2-dependent migration of oligodendrocyte precursors in the
developing optic nerve. Mol. Cell. Neurosci. 33, 2–14 (2006).
18. Maysami, S. et al. Modulation of rat oligodendrocyte precursor cells by the
chemokine CXCL12. Neuroreport 17, 1187–1190 (2006).
19. Tsai, H. H. et al. The chemokine receptor CXCR2 controls positioning of
oligodendrocyte precursors in developing spinal cord by arresting their migration.
Cell 110, 373–383 (2002).
97
20. Spassky, N. et al. Directional guidance of oligodendroglial migration by class 3
semaphorins and netrin-1. J. Neurosci. 22, 5992–6004 (2002).
21. Taniguchi, Y. et al. Sema4D deficiency results in an increase in the number of
oligodendrocytes in healthy and injured mouse brains. J. Neurosci. Res. 87,
2833–2841 (2009).
22. Armendáriz, B. G. et al. Expression of Semaphorin 4F in neurons and brain
oligodendrocytes and the regulation of oligodendrocyte precursor migration in the
optic nerve. Mol. Cell. Neurosci. 49, 54–67 (2012).
23. Lang, J. et al. Adenomatous Polyposis Coli Regulates Oligodendroglial
Development. J. Neurosci. 33, 3113–3130 (2013).
24. Emery, B. Regulation of Oligodendrocyte Differentiation and Myelination. 330,
779–782 (2010).
25. Chen, Y. et al. The oligodendrocyte-specific G protein-coupled receptor GPR17 is
a cell-intrinsic timer of myelination. Nat. Neurosci. 12, 1398–1406 (2009).
26. Bergstrom-tyrberg, U., Berglund, E. O. & Ranscht, B. Contactin-1 regulates
myelination and nodal / paranodal domain organization in the central nervous
system. (2014). doi:10.1073/pnas.1313769110
27. Wang, S. et al. Notch receptor activation inhibits oligodendrocyte differentiation.
Neuron 21, 63–75 (1998).
28. Hu, Q. D. et al. F3/contactin acts as a functional ligand for notch during
oligodendrocyte maturation. Cell 115, 163–175 (2003).
29. Ye, P., Carson, J. & D’Ercole, A. J. In vivo actions of insulin-like growth factor-I
(IGF-I) on brain myelination: studies of IGF-I and IGF binding protein-1 (IGFBP-1)
98
transgenic mice. J. Neurosci. 15, 7344–56 (1995).
30. Gibson, E. M. et al. Neuronal Activity Promotes Oligodendrogenesis and Adaptive
Myelination in the Mammalian Brain. 344, 1–27 (2014).
31. OShea, J. J., Gadina, M. & Schreiber, R. D. Cytokine signaling in 2002: New
surprises in the Jak/Stat pathway. Cell 109, 121–131 (2002).
32. Zhang, J.-M. & An, J. Cytokines, Inflammation and Pain. Int Anesth. Clin. 69,
482–489 (2009).
33. Muller-Newen, G. The Cytokine Receptor gp130: Faithfully Promiscuous. Sci.
Signal. 2003, pe40-pe40 (2003).
34. Taga, T. & Kishimoto, T. gp130 AND THE INTERLEUKIN-6 FAMILY OF
CYTOKINES. Annu. Rev. Immunol. 15, 797–819 (1997).
35. White, U. A. & Stephens, J. M. The gp130 receptor cytokine family: regulators of
adipocyte development and function. Curr. Pharm. Des. 17, 340–6 (2011).
36. Levy, D. E., Lee, C., Levy, D. E. & Lee, C. What does Stat3 do ? Find the latest
version : What does Stat3 do ? 109, 1143–1148 (2002).
37. Silver, J. S. & Hunter, C. A. Gp130 At the Nexus of Inflammation, Autoimmunity,
and Cancer. J. Leukoc. Biol. 88, 1145–1156 (2010).
38. Gregg, C. CNTF/LIF/gp130 receptor complex signaling maintains a VZ precursor
differentiation gradient in the developing ventral forebrain. Development 132,
565–578 (2005).
39. Bonaguidi, M. A. LIF and BMP signaling generate separate and discrete types of
GFAP-expressing cells. Development 132, 5503–5514 (2005).
40. Barnabé-Heider, F. et al. Evidence that embryonic neurons regulate the onset of
99
cortical gliogenesis via cardiotrophin-1. Neuron 48, 253–265 (2005).
41. Drogemuller, K. et al. Astrocyte gp130 Expression Is Critical for the Control of
Toxoplasma Encephalitis. J. Immunol. 181, 2683–2693 (2008).
42. Deverman, B. E. & Patterson, P. H. Cytokines and CNS Development. Neuron 64,
61–78 (2009).
43. Fischer, A. J., Schmidt, M., Omar, G. & Reh, T. A. BMP4 and CNTF are
neuroprotective and suppress damage-induced proliferation of Müller glia in the
retina. Mol. Cell. Neurosci. 27, 531–542 (2004).
44. Todd, L., Squires, N., Suarez, L. & Fischer, A. J. Jak/Stat signaling regulates the
proliferation and neurogenic potential of Müller glia-derived progenitor cells in the
avian retina. Sci. Rep. 6, 1–16 (2016).
45. Ishibashi, T., Lee, P. R., Baba, H. & Fields, R. D. Leukemia Inhibitory Factor
Regulates the Timing of Oligodendrocyte Development and Myelination in the
Postnatal Optic Nerve. 3355, 3343–3355 (2009).
46. Barres, B. a, Schmid, R., Sendnter, M. & Raff, M. C. Multiple extracellular signals
are required for long-term oligodendrocyte survival. Development 118, 283–295
(1993).
47. Barres, B. A. et al. Ciliary neurotrophic factor enhances the rate of
oligodendrocyte generation. Mol. Cell. Neurosci. 8, 146–156 (1996).
48. Mayer, M., Bhakoo, K. & Noble, M. Ciliary neurotrophic factor and leukemia
inhibitory factor promote the generation, maturation and survival of
oligodendrocytes in vitro. Development 120, 143–153 (1994).
49. Zhang, Y. et al. Interleukin-11 Potentiates Oligodendrocyte Survival and
100
Maturation, and Myelin Formation. J. Neurosci. 26, 12174–12185 (2006).
50. Rao, M. Transmembrane Protein GDE2 Induces Motor Neuron Differentiation in
Vivo. Science (80-. ). 309, 2212–2215 (2005).
51. Park, S. et al. GDE2 promotes neurogenesis by glycosylphosphatidylinositol-
anchor cleavage of RECK. Science (80-. ). 339, 324–328 (2013).
52. Yan, Y., Sabharwal, P., Rao, M. & Sockanathan, S. The Antioxidant Enzyme
Prdx1 Controls Neuronal Differentiation by Thiol-Redox-Dependent Activation of
GDE2. Cell 138, 1209–1221 (2009).
53. Cave, C. et al. GDE2 is essential for neuronal survival in the postnatal
mammalian spinal cord. Mol. Neurodegener. 12, 1–20 (2017).
54. Veen, M. Van, Mans, L. A., Matas-rico, E., Pelt, J. Van & Perrakis, A. International
Journal of Biochemistry and Cell Biology Glycerophosphodiesterase GDE2 /
GDPD5 a ff ects pancreas di ff erentiation in zebra fi sh. 94, 71–78 (2018).
55. Matas-rico, E. et al. Glycerophosphodiesterase GDE2 Promotes Neuroblastoma
Differentiation through Glypican Release and Is a Marker of Clinical Outcome
Article Glycerophosphodiesterase GDE2 Promotes Neuroblastoma Differentiation
through Glypican Release and Is a Marker of Clin. 548–562 (2016).
doi:10.1016/j.ccell.2016.08.016
56. Corda, D. et al. The developmentally regulated osteoblast phosphodiesterase
GDE3 is glycerophosphoinositol-specific and modulates cell growth. J. Biol.
Chem. 284, 24848–24856 (2009).
57. Van Veen, M. et al. Negative regulation of urokinase receptor activity by a GPI-
specific phospholipase C in breast cancer cells. Elife 6, 1–20 (2017).
101
58. Müller, G. A. The release of glycosylphosphatidylinositol-anchored proteins from
the cell surface. Arch. Biochem. Biophys. 656, 1–18 (2018).
59. György, B. et al. Membrane vesicles, current state-of-the-art: Emerging role of
extracellular vesicles. Cell. Mol. Life Sci. 68, 2667–2688 (2011).
60. Rajendran, L. et al. Emerging Roles of Extracellular Vesicles in the Nervous
System. J. Neurosci. 34, 15482–15489 (2014).
61. Saman, S. et al. Exosome-associated tau is secreted in tauopathy models and is
selectively phosphorylated in cerebrospinal fluid in early Alzheimer disease. J.
Biol. Chem. 287, 3842–3849 (2012).
62. Aguzzi, A. & Rajendran, L. The Transcellular Spread of Cytosolic Amyloids,
Prions, and Prionoids. Neuron 64, 783–790 (2009).
63. Witwer, K. W. et al. Standardization of sample collection, isolation and analysis
methods in extracellular vesicle research. J. Extracell. Vesicles 2, 0–25 (2013).
64. Akers, J. C., Gonda, D., Kim, R., Carter, B. S. & Chen, C. C. Biogenesis of
extracellular vesicles (EV): Exosomes, microvesicles, retrovirus-like vesicles, and
apoptotic bodies. J. Neurooncol. 113, 1–11 (2013).
65. Colombo, M., Raposo, G. & Théry, C. Biogenesis, Secretion, and Intercellular
Interactions of Exosomes and Other Extracellular Vesicles. Annu. Rev. Cell Dev.
Biol. 30, 255–289 (2014).
66. Babst, M. MVB vesicle formation: ESCRT-dependent, ESCRT-independent and
everything in between. Curr. Opin. Cell Biol. 23, 452–457 (2011).
67. Abels, E. R. & Breakefield, X. O. Introduction to Extracellular Vesicles:
Biogenesis, RNA Cargo Selection, Content, Release, and Uptake. Cell. Mol.
102
Neurobiol. 36, 301–312 (2016).
68. Henne, W. M., Stenmark, H. & Emr, S. D. Sculpting ESCRT Pathway. 1–12
(2013). doi:10.1101/cshperspect.a016766
69. McCullough, J. et al. Structure and membrane remodeling activity of ESCRT-III
helical polymers. Science (80-. ). 350, 1548–1551 (2015).
70. Baietti, M. F. et al. Syndecan-syntenin-ALIX regulates the biogenesis of
exosomes. Nat. Cell Biol. 14, 677–685 (2012).
71. Trajkovic, K. et al. Ceramide Triggers Budding of Exosome Vesicles into
Multivesicular Endosomes. Science (80-. ). 319, 1244–1247 (2007).
72. Blanc, L. & Vidal, M. New insights into the function of Rab GTPases in the context
of exosomal secretion. Small GTPases 9, 95–106 (2018).
73. Ostrowski, M. et al. Rab27a and Rab27b control different steps of the exosome
secretion pathway. Nat. Cell Biol. 12, 19–30 (2010).
74. Parolini, I. et al. Microenvironmental pH is a key factor for exosome traffic in tumor
cells. J. Biol. Chem. 284, 34211–34222 (2009).
75. Nanbo, A., Kawanishi, E., Yoshida, R. & Yoshiyama, H. Exosomes Derived from
Epstein-Barr Virus-Infected Cells Are Internalized via Caveola-Dependent
Endocytosis and Promote Phenotypic Modulation in Target Cells. J. Virol. 87,
10334–10347 (2013).
76. Skotland, T., Sandvig, K. & Llorente, A. Lipids in exosomes: Current knowledge
and the way forward. Prog. Lipid Res. 66, 30–41 (2017).
77. Zhou, J. et al. Retinal progenitor cells release extracellular vesicles containing
developmental transcription factors, microRNA and membrane proteins. Sci. Rep.
103
8, 1–15 (2018).
78. Atai, N. A. et al. Heparin blocks transfer of extracellular vesicles between donor
and recipient cells. J. Neurooncol. 115, 343–351 (2013).
79. Christianson, H. C., Svensson, K. J., van Kuppevelt, T. H., Li, J.-P. & Belting, M.
Cancer cell exosomes depend on cell-surface heparan sulfate proteoglycans for
their internalization and functional activity. Proc. Natl. Acad. Sci. 110, 17380–
17385 (2013).
80. Cossetti, C. et al. Extracellular vesicles from neural stem cells transfer IFN-γ via
Ifngr1 to activate Stat1 signaling in target cells. Mol. Cell 56, 193–204 (2014).
81. Kooyman, D. L. et al. In vivo transfer of GPI-linked complement restriction factors
from erythrocytes to the endothelium. Science (80-. ). 269, 89–92 (1995).
82. Anderson, S. M., Yu, G., Giattina, M. & Miller, J. L. Intercellular transfer of a
glycosylphosphatidylinositol (GPI)-linked protein: release and uptake of CD4-GPI
from recombinant adeno-associated virus-transduced HeLa cells. Proc. Natl.
Acad. Sci. U. S. A. 93, 5894–8 (1996).
83. Silbereis, J. C., Huang, E. J., Back, S. A. & Rowitch, D. H. Towards improved
animal models of neonatal white matter injury associated with cerebral palsy. Dis.
Model. Mech. 3, 678–688 (2010).
84. Pasquin, S., Sharma, M. & Gauchat, J. F. Ciliary neurotrophic factor (CNTF): New
facets of an old molecule for treating neurodegenerative and metabolic syndrome
pathologies. Cytokine Growth Factor Rev. 26, 507–515 (2015).
85. Martynoga, B., Morrison, H., Price, D. J. & Mason, J. O. Foxg1 is required for
specification of ventral telencephalon and region-specific regulation of dorsal
104
telencephalic precursor proliferation and apoptosis. Dev. Biol. 283, 113–127
(2005).
86. Wu, H. et al. Cellular resolution maps of X-chromosome inactivation: implications
for neural development, function, and disease Hao. 81, 103–119 (2014).
87. Consortium, T. G. O. The Gene Ontology project in 2008. Nucleic Acids Res. 44,
273–290 (2007).
88. Cahoy, J. D. et al. A Transcriptome Database for Astrocytes, Neurons, and
Oligodendrocytes: A New Resource for Understanding Brain Development and
Function. J. Neurosci. 28, 264–278 (2008).
89. Stankoff, B. et al. Ciliary neurotrophic factor (CNTF) enhances myelin formation: a
novel role for CNTF and CNTF-related molecules. J. Neurosci. 22, 9221–7
(2002).
90. Emery, B. & Dugas, J. C. Purification of oligodendrocyte lineage cells from mouse
cortices by immunopanning. Cold Spring Harb. Protoc. 2013, 854–868 (2013).
91. Chevalier, S. et al. Interleukin-6 family of cytokines induced activation of different
functional sites expressed by gp130 transducing protein. J. Biol. Chem. 271,
14764–14772 (1996).
92. Lötvalll, J. et al. Minimal experimental requirements for definition of extracellular
vesicles and their functions: A position statement from the International Society
for Extracellular Vesicles. J. Extracell. Vesicles 3, 1–6 (2014).
93. Di Marco, A. et al. Agonistic and antagonistic variants of ciliary neurotrophic factor
(CNTF) reveal functional differences between membrane-bound and soluble
CNTF ??-receptor. J. Biol. Chem. 272, 23069–23075 (1997).
105
94. Davis, S. et al. Released form of CNTF receptor ( alpha ) component as a solubl.
Science (80-. ). 259, 1736–1739 (1993).
95. Wang, H. et al. MiR-696 Regulates C2C12 Cell Proliferation and Differentiation by
Targeting CNTFR α. 13, (2017).
96. Hughes, E. G., Kang, S. H., Fukaya, M. & Bergles, D. E. HHS Public Access. 16,
668–676 (2013).
97. Port, M. D., Gibson, R. M. & Nathanson, N. M. Differential stimulation-induced
receptor localization in lipid rafts for interleukin-6 family cytokines signaling
through the gp130/leukemia inhibitory factor receptor complex. J. Neurochem.
101, 782–793 (2007).
98. Tripathi, R. B. & McTigue, D. M. Chronically increased ciliary neurotrophic factor
and fibroblast growth factor-2 expression after spinal contusion in rats. J. Comp.
Neurol. 510, 129–144 (2008).
99. Ernsberger, U., Sendtner, M., D-, N. & Republic, M. F. Proliferation and
Differentiation of Embryonic Chick Sympathetic Neurons : Effects of Ciliary
Neurotrophic Factor. 2, 1275–1284 (1989).
100. Johnson, R. W., White, J. D., Walker, E. C., Martin, T. J. & Sims, N. A. Myokines
(muscle-derived cytokines and chemokines) including ciliary neurotrophic factor
(CNTF) inhibit osteoblast differentiation. Bone 64, 47–56 (2014).
101. Dewitt, J., Pappas, A. & Nishi, R. Ciliary neurotrophic factor reduces proliferation
and promotes differentiation of TH-MYCN transformed sympathoadrenal
progenitors. Dev. Neurosci. 36, 422–431 (2015).
102. Sun, Y. & Maxwell, G. D. Ciliary neurotrophic factor (CNTF) has a dose-
106
dependent biphasic effect on the number of adrenergic cells which develop in
avian trunk neural crest cultures. Neurosci. Lett. 165, 1–4 (1994).
103. Laulagnier, K. et al. PLD2 is enriched on exosomes and its activity is correlated to
the release of exosomes. FEBS Lett. 572, 11–14 (2004).
104. Bahrini, I., Song, J. H., Diez, D. & Hanayama, R. Neuronal exosomes facilitate
synaptic pruning by up-regulating complement factors in microglia. Sci. Rep. 5, 1–
8 (2015).
105. Wang, T. et al. Hypoxia-inducible factors and RAB22A mediate formation of
microvesicles that stimulate breast cancer invasion and metastasis. Proc. Natl.
Acad. Sci. 111, E3234–E3242 (2014).
106. Xie, Y. et al. GPS-Lipid: A robust tool for the prediction of multiple lipid
modification sites. Sci. Rep. 6, 1–9 (2016).
107. Levental, I., Lingwood, D., Grzybek, M., Coskun, U. & Simons, K. Palmitoylation
regulates raft affinity for the majority of integral raft proteins. Proc. Natl. Acad. Sci.
107, 22050–22054 (2010).
108. Simons, K. & Sampaio, J. Membrane organization and lipid rafts. Cold Spring
Harb Perspect Biol 1–18 (2011).
109. Paulick, M. G. & Bertozzi, C. R. The glycosylphosphatidylinositol anchor: A
complex membrane-anchoring structure for proteins. Biochemistry 47, 6991–7000
(2008).
110. Bate, C., Nolan, W. & Williams, A. Sialic acid on the glycosylphosphatidylinositol
anchor regulates PrP-mediated cell signaling and prion formation. J. Biol. Chem.
291, 160–170 (2016).
107
Biography
Mateusz Dobrowolski was born in 1990 in Swidnica, Poland.
Mateusz did his undergraduate work at Temple University in Philadelphia, PA where he majored in Biochemistry. During his undergraduate studies, he worked under the supervision of Dr. Karen Palter where he investigated the loss of sialylation in
Drosophila.
In 2012, Mateusz began his PhD at Johns Hopkins University School of Medicine under the mentorship of Dr. Shanthini Sockanathan.
108